journals/urban...2 urban transport journal editorial board dr. sanjay gupta (chairman) school of...

1

URBAN TRANSPORT

Journal

Vol 13 No.1 September 2014

INSTITUTE OF URBAN TRASNPORT (INDIA)

2

URBAN TRANSPORT

Journal

Editorial Board

Dr. Sanjay Gupta (Chairman)

School of Planning & Architecture, Delhi

Dr.Geetam Tiwari Shri Piyush Kansal

Indian Institute of Technology, Delhi RITES Ltd, Gurgaon

Dr. Ashish Verma Dr.Pawan Kumar

Indian Institute of Science, Bengaluru Town & Country Planning Organization, Delhi

Shri C.L.Kaul(Convenor)

All communication pertaining to submission of papers for publication in the journal may be sent by e-mail at

the following address:

Executive Secretary

Institute of Urban Transport (India)

1st Floor, AnandVihar Metro Station Building,

(Entry adjacent to Gate No 1)Delhi - 110 092

Phones: (91) 11 66578730 & 40 (D) (91) 11 66578700-709 (10 lines) Ext. - 730, 740

Telefax:91) 11 66578733

E-mail: [email protected]

Website: www.iutindia.org

All rights reserved

Views expressed in the papers in this Journal are those of the authors and not necessarily the views of the

Institute.

http://www.iutindia.org/

URBAN TRANSPORT

JOURNAL

Vol.13 No.1, Sept 2014

From the Editor’s Desk

The present issue of the journal comprises of eight technical papers presenting diverse spectrum of

research themes such as global walkability index, event day effect on pedestrian behavior,

effect of lane friction on speed of NMV’s, service quality determinants for public transport

and use intention, framework for estimating carbon footprint of commuters, modeling mode

choice behaviour of commuters, roundabout entry capacity model, framework for development

of advanced traveler information system and institutional & financial strengthening of

intermediate public transport services in Indian cities.

The first paper on “Application of Global Walkability Index (GWI: Case study of Bangalore,

India” is an attempt to demonstrate an application of nine point street rating tool developed

by the World Bank and Clean Air Initiative (CAI) Asia to four streets with major pedestrian

footfall in the metropolitan city of Bangalore, India. It highlights the advantages of having

the tool in a smartphone application format with connectivity to a global crowd map for

application by professionals working towards street improvement designs and aid creating a

global database of knowledge.

The second paper on “Event Day Effect on Pedestrian Characteristics for CBD Street of

Indian Metropolitan City” focusses on the pedestrian flow behaviour on normal week day and

event day captured t hr ou gh video recording of pedestrian movement in CBD area of

Vadodara city. The paper highlights the variations in pedestrian flow behavior during normal

and religious event day in terms of volume, speed, density, space and level of service of

pedestrian flow in both conditions. The author reiterates the need for TSM actions to improve

pedestrian movement on an event day.

The third paper on “ Effect of Lane Friction on Speed of Non-Motorized Vehicles” is an

interesting attempt on the NMV (bicycle and cycle rickshaw together) flows on urban roads

based on an empirical study carried out in Roorkee city for three roads, all of two-lanes with

either one-directional or two-direction traffic. The paper concludes that the speeds were

influenced by share of NMV flows. In particular on a road segment with no friction from

opposite direction, the NMV speed is reduced with an increase in total flow; on segment with

friction from opposite direction it is reduced with increase in opposing direction flows, while

on segment with all frictions it is reduced with NMV, total and opposing direction flows

The fourth paper on “Service Quality Determinants for Public Transport and Use Intention: A

Study of Commuters and Non- Commuters in India” The paper attempts to address the prime

objective of identifying the service quality of determinants for commuters and non-commuters

and their influence on use intention. This paper is based on empirical studies conducted in the

cities of Delhi, Mumbai, Allahabad and Jabalpur amongst the commuters and non-commuters.

It was found that both the user groups gave weightage to availability and tangibles factors as

important service quality determinants though in different order along with empathy,

responsiveness for commuters and safety, integration for non-commuters respectively.

The fifth paper on “Generic Framework for Estimating Carbon Footprint of Commuting with

Public Transport Modes” demonstrates a method for estimating the carbon footprint of

commuting and apply it to the public transport systems existing in Delhi. It is based on an

empirical study carried out in Delhi of the transit commuters of available modes to estimate

the carbon foot prints for different mode- combinational trips (trip profile including access,

egress and main line haul mode). The authors reiterate that carbon footprints assessment has

the potential to provide insights into the potential impact of different policies.

The sixth paper on “Modeling Mode Choice Behaviour and Estimating Value of Travel Time

of Commuters in Delhi” reiterates that mode choice forms an integral part of travel demand

modeling as it gives a complete insight to the mode choice preferences of the commuters. In

their paper the authors attempts to model the mode choice of commuters in Delhi based on

discrete multinomial logit model. The paper is based on an empirical study of various

localities of Delhi and uses thirteen explanatory variables for modeling mode choice behavior.

It also quantifies the value of travel time separately for motorized and non motorized mode of

commuters.

The seventh paper on “Selection of Roundabout Entry Capacity Model for Indian Condition”

emphasizes that evaluating the capacity of roundabout is an important element in the planning

and design of such facilities. In this paper an empirical approach using regression analysis was

used to develop a roundabout entry-capacity model for Indian conditions. It was observed that

the entry capacity of an approach of a roundabout is dependent on the circulating flow in front

of that approach Also the critical gap and follow-up time values recommended in HCM (2010)

are not applicable to Indian conditions .The paper concludes that the capacity based on U.S.

model gave results close to that of field entry capacity model.

The eight paper on “Framework for Development of Advanced Traveler Information System:

A Case Study for Chandigarh City” presents a comprehensive framework comprising of

system architecture, development methodology and salient features of a GIS based ATIS for

city of Chandigarh City, India. The author emphasizes that the suggested system is able to

provide the information about the basic facilities of the city and help the users in planning and

decision making about their trips by providing shortest routes, nearest facilities and bus routes.

The last paper on “Institutional and Financial Strengthening of Intermediate Public Transport

Services in Indian Cities” focuses on the Intermediate Public Transport like 3 wheelers, auto

rickshaws, tempos and tata magic that caters the daily urban trips in Indian cities. This paper

identifies the key challenges faced by this sector and provides recommendations for addressing

these challenges.

Dr. Sanjay Gupta

URBAN TRANSPORT

Journal

Volume 13 No.1 September 2014

Contents

Application of Global Walkability Index (GWI): Case Study Bangalore, India

Neelakshi Joshi, Prof. R. Shankar and Prof. Dr. Ing Helmut Bott

1

Event Day Effect on Pedestrian Characteristics for CBD Street of Indian Metropolitan

City

Hardik S Sukhadia, Sanjay M Dave, Jiten Shah, Dipak Rathva

14

Effect of Lane Friction on Speed of Non-Motorized Vehicles

Prasham Khadaiya and Rajat Rastogi

26

Service Quality Determinants for Public Transport and Use Intention: A Study of

Commuters and Non-Commuters in India

Dr. Vibhuti Tripathi and Gunjan Nema

39

Generic Framework for Estimating Carbon Footprint of Commuting with Public

Transport Modes

Kirti Bhandari, Mukti Advani, Purnima Parida

57

Modeling Mode Choice Behaviour and Estimating Value of Travel Time of Commuters

in Delhi

Minal and Ch.Ravi Sekhar

67

Selection of Roundabout Entry Capacity Model for Indian Condition

Abdullah Ahmad, Srinath Mahesh and Rajat Rastogi

78

Framework for Development of Advanced Traveler Information System: A Case Study

for Chandigarh City

Bhupendra Singh, Ankit Gupta, Sanjeev Suman

87

Institutional and Financial Strengthening of Intermediate Public Transport Services in

Indian Cities

Anindita Ghosh and Kanika Kalra

96

URBAN TRANSPORT

JOURNAL


* Indian Institute of Technology, Roorkee and University of Stuttgart

[email protected]

APPLICATION OF GLOBAL WALKABILITY INDEX (GWI): CASE

STUDY BANGALORE, INDIA

Neelakshi Joshi, Prof. R. Shankar and Prof. Dr. Ing Helmut Bott*

Abstract: Global Walkability Index (GWI) is a nine point street rating tool developed by the World Bank and

Clean Air Initiative (CAI) Asia. This paper applies GWI to four streets with major pedestrian footfall in the

metropolitan city of Bangalore, India. The possibility of adding a tenth criterion i.e. of ‘Environment Quality’ is

explored. Data collected is further analyzed and is found to generate a score of 20.25 out of 50 for the city,

which is low. Furthermore, the advantages of having the tool in a smartphone application format with

connectivity to a global crowdmap are discussed. This paper aims to encourage those working towards street

improvement designs to widely apply this emerging tool and help create a global database of knowledge.

Keywords: GlobalWalkability IndexWalkscore, Pedestrians, Mobility

1.0 INTRODUCTION

Indian cities, in recent years, have

experienced a steady shift towards private

vehicular ownership. As per Census of India

2011, urban car ownership stands at 9.7% and

two-wheeler ownership at 35.2%. Though

these are still low compared to global

standards, their increasing numbers on Indian

roads are leading to major congestion

problems and increase in air and sound

pollution. In Bangalore, transport is

responsible for 42% of all the air pollution

(TERI, 2010). Furthermore, pedestrians’ share

in road accidents is also on the rise, being as

high as 44% in Bangalore (MoUD, 2008).

Municipal funding primarily caters to road

widening and flyover projects, without

adequate improvement of non-motorized

facilities. This is evident from the transport

projects undertaken under Jawaharlal Nehru

National Urban Renewal Mission (JNNURM)

between 2006-10. Around 56% of the

transport projects have been either for road

widening or flyover construction, 33% for

public transport and 5% for improving parking

facilities. The remaining 5% is categorized as

‘others’ and pedestrians may or may not be

included in this segment (IIHS, 2011). This

apathy towards pedestrians coupled with an

unsafe and unpleasant walking environment

further encourages people to shift to motorized

modes.

Walkability is a measure of the ease of

walking from one place to another. Definitions

differ as contexts change. Some attribute it to

the nature of land use: “The extent to which

the built environment is friendly to the

presence of people living, shopping, visiting,

enjoying or spending time in an area” (Abbey,

2005) while others attribute it to physical

infrastructure, comfort and safety from crime

(Krembeck et.al, 2006).

It is true that mixed land use, facilitating

ease of walking to work and recreation, is an

essential tool for enhancing walk trips.

However, in the context of India, most streets

are intrinsically mixed use and applying mixed

land use parameters like ‘Walkscore’ generate

good results. Walkscore is an internet based

service that analyzes streets geospatially based

on the distance of various amenities from the

said location.High scores are generated for

mixed use areas where people can walk to

access various facilities like shops, schools,

restaurants. However, the real problem of bad

walking environment, lack of comfort and

safety can be gauged well by indices that take

a more microscopic approach.


2

2.0 GLOBAL WALKABILITY

INDEX (GWI)

Global Walkability Index (GWI) was

developed by Holly Krambeck and Jitendra

Shah (Krembeck et.al, 2006). It rates streets

and the walking environment based on nine

predefined criteria. These criteria cover

aspects of safety, security and comfort (see

Table 1). Streets are rated on a scale of 0-5, 0

being a complete absence of the said

parameter and 5 indicating excellent

conditions. Length of the street and pedestrian

count further contribute towards developing a

rating. This aims to quantify the problems of

bad pedestrian environment and furthermore

develop a universal scale for measuring and

comparing these.

Clean Air Initiative (CAI), Asia conducted

field walkability surveys, following the GWI

parameters in 13 Asian cities namely Cebu

(Philippines), Colombo (Sri Lanka), Davao

(Philippines), Ha Noi (Viet Nam), Ho Chi

Minh City (Viet Nam), Hong Kong, China

(People’s Republic of China), Jakarta

(Indonesia), Karachi (Pakistan), Kathmandu

(Nepal), Kota (India), Lanzhou (PRC), Manila

(Philippines), and Ulaanbaatar (Mongolia)

(Leather, 2011). It eliminated the variables of

pedestrian count and length of street surveyed

as including these often resulted in streets with

high pedestrian counts but bad infrastructure to

get high scores.

Besides the nine point criteria adopted by

CAI, the authors further explored adding a

tenth criteria of ‘Environment Quality’

measured in terms of air quality and sound

pollution levels as pedestrians are directly

exposed to this during their walk trips

Table 1: Global Walkability Index (GWI) Parameters

No. Parameter Score Description

1 Walking Path

Modal

Conflict

1 Significant conflict that makes walking impossible

2 Walking possible, but dangerous and inconvenient

3 Some conflict – walking is possible, but not convenient

4 Minimal conflict, mostly between non-motorized modes

5 No conflict between pedestrians and other modes

2 Availability

Of Walking

Paths

1 Pedestrian walkways required but not available

2 Available but highly congested, bad maintenance

3 Available but congested, needs better maintenance

4 Available, sometimes congested, well maintained

5 Walkways not required: people can safely walk on roads

3 Availability

Of Crossings

1 Controlled crossings is > 500m and average speed is

high

2 Controlled crossings is between 500-300m and average

speed is around 40 Kmph


speed is 20-40 Kmph


speed is 20-40 Kmph


3


5 Vehicles and pedestrians coexist safely

4 Grade

Crossing

Safety

1 High probability of accident; very high crossing time

2 Dangerous- pedestrian faces some risk of being hurt by

other modes and crossing time is high

3 Difficult to ascertain dangers posed to pedestrians but

the time available for crossing is less

4 Safe –exposure time is less and time available for

crossing more

5 Very safe – other modes present no danger to

pedestrians

5 Motorist

Behavior

1 High traffic disrespect to pedestrians

2 Traffic disrespect and rarely pedestrians get priority

3 Motorists sometimes yield

4 Obey laws and sometimes yield to pedestrians

5 Obey traffic laws and almost always yield to pedestrians

6 Amenities 1 No Amenities

2 Little amenities at some locations

3 Limited number of provisions for pedestrians

4 Good amenities for major length

5 Excellent amenities such as lighting, cover from sun and

rain making walking a pleasant experience

7 Disability

Infrastructure

1 No infrastructure for disabled people is available

2 Limited infrastructure but not in usable condition

3 Present but in poor condition and not well placed

4 Present, in good condition, but poorly placed

5 Present, in good condition, and well placed

8 Obstructions 1 Infrastructure is completely blocked by obstructions

2 Significantly inconvenienced. Effective width <1m

3 Mildly inconvenienced; effective width is < or = 1 meter

4 Minor inconvenience. Effective width is > 1m

5 There are no obstructions

9 Security from

Crime

1 Very dangerous–pedestrians highly susceptible to crime

2 Dangerous – pedestrians are at some risk of crime


4


3 Difficult to ascertain perceived degree of security

4 Secure – pedestrians at minimal crime risk

5 Very secure – pedestrians at virtually no risk of crime

10 Environment

Quality*

1 Pedestrian exposed to very high air and sound pollution

2 Air pollution high and sound between 80-90db

3 Air pollution moderate, sound between 60-80db

4 No perceptible air pollution, sound between 50-60bd

5 Pleasant air quality, Sound <50db

Source: Krambeck& Shah, 2006 and *Author

CAI also developed a smartphone app

called ‘Walkability’ making it easy to conduct

field surveys using smartphones (see Image 1).

The rating criteria has been simplified and

presented in an easy to understand format. The

idea is to enable citizens to easily rate their

surrounding streets. The ratings posted on the

app appear on a global crowd map which

automatically generates city and country score

based on these

(http://www.dotzoo.net/walkability/).

3.0 METHODOLOGY

Four streets in the metropolitan city of

Bangalore were identified for conducting the

walkability survey (see Figure 2). High

pedestrian count in these streets was the

primary criteria for selection (see Table 2).

These streets were also found to have heavy

motor traffic gauged through high values of

Passenger Car Units (PCU). Incidentally, all

four are primarily commercial streets with

active shop frontages. Walkability surveys

were conducted between 11:00 am - 12:00

noon on four subsequent Sundays between

26.5.2013 to 16.6.2013. Sunday noon has been

identified as the peak for pedestrians in these

areas (RITES, 2011).

Figure 1: Streets identified for study,

Bangalore

Figure 1: Screenshot of Walkability App

Source: tps://itunes.apple.com/us/app/cai-asia-

walka-bility-app/id567092795?mt=8


5

Table 2: Streets identified for study, Bangalore

Sr. Name of Street

Length of

street

surveyed (m)

Pedestrian

Count/Ho

ur

PCU/Hour One way

1. 9th Main, Jayanagar 240m 5,797 89,376 Yes

2. Brigade Road 355m 5,198 59,832 Yes

3. Sampige Road,

Malleshwaram 404m 3,110* 68,045 Yes

4. Gandhi Bazaar Road,

Baswangudi 660m 2,578 1,02,678 No

Source: Rites India and *Author

All ten criteria described in Table 1 were

taken into consideration for rating streets.

Each criteria was scored on a scale of 0 to 5, 0

being the worst and 5 being the best situation.

Each street was rated overall out of 50 points.

The average of all four streets suggests the city

score. Sound pollution levels were measured

employing ‘Noise Meter’ app that helps record

the decibel levels in an area. Air pollution

levels are difficult to gauge as current

smartphone do not have sensors to measure

this. Hence, air pollution was qualitatively

analyzed by the author based on a comparative

analysis of the four streets. Results were

tabulated and analyzed to develop and overall

city score. Furthermore, the results generated

using GWI were compared against

‘Walkscore’ for the same streets. ‘Walkscore’

for a particular street was generated by

entering its location on www.walkscore.com.

Scores are generated automatically by a

geospatial analysis of distance of various

amenities like restaurants, schools, offices etc

from the said location.

4.0 LIMITATIONS

GWI primarily relies on the perception

of the surveyor making it a qualitative tool

for analysis. Despite providing a scale for

rating various aspects of the walking

environment, there is scope for individual

interpretation and variation. Bias of the

surveyor can further impact the rating. The

10th criteria on ‘Environment Quality’

suggests using in-built sensors of a smart

phone to quantify noise and air pollution.

Training surveyors and providing

interpretation manuals for online users are

some steps taken by Clean Air Asia to

generate consistent rating.

5.0 SURVEY

Surveys were conducted rating streets as

per the GWI parameters. Furthermore air and

noise exposure levels were also captured.

Photographs were taken to capture certain

aspects of the survey better. Following is a

detailed account of each road:

5.1 9th

Main Road, Jayanagar

9th Main Road is an active commercial

street in Bangalore located in the residential

neighbourhood of Jayanagar. It records the

highest pedestrian counts. Though some effort

has gone into separating pedestrians from

motor vehicles, providing benches, installing

dustbins and street lighting the streets are

unsafe to cross as cars freely pass through the

area and do not yield to pedestrians. Also,

there is encroachment by shopkeepers and

vendors forcing pedestrians to walk on the

road with cars. Noise levels are high owing to

constant honking by cars to alert pedestrians.


6

Table 3: Walkability Survey, 9th Main Road, Jayanagar

Source: Author

5.2 Brigade Road

Brigade Road is the main commercial street

in Bangalore city. It is located in the heart of

the central business district. Pedestrians on this

street are restricted to footpaths that are clearly

inadequate to hold such large pedestrian

volumes. The central spine is for cars making

it difficult to cross the road. Air and sound

pollution from the cars and motor cycles

further affect the riding environment.

Shopkeepers often encroach a part of the

footpath. There are no places to sit and relax.

Furthermore, because of pedestrians packed on

the footpaths, pick pocketing and harassment

of women is common.

Criteria Remark Score

Walking Path

Modal Conflict

Pedestrian pathways exist but people end up walking on road due to

encroachment (see Figure 3a) 2

Availability of

Walking Paths

3.8 - 4.2 m wide on both sides

Encroached by local vendors

Trees, dustbins and lampposts also obstruct walking

3

Availability of

Crossings

No defined crossing except at junctions (250m)

People jaywalk to cross

Zebra crossings obstructed by railings (see Figure 3b)

2

Grade Crossing

Safety

Traffic is not controlled hence pedestrian has to negotiate the road

himself 2

Motorist

Behavior

Persistent honking

No priority to pedestrians 2

Amenities Physical separation from main road

Street lights and dustbins at 100m (see Figure 3c)

Benches in open areas

3

Disability

Infrastructure

Footpath heights are not negotiable by wheelchair

Frequent obstructions 0

Obstructions Encroachments from shops

Existing physical barriers like lamp posts, dustbins and trees 2

Security from

Crime

Vibrant mixed use

Improvement in lighting is needed 3

Walking

Environment

Noise level at 82 dB

Air pollution levels high 2

Total 21/50

Figure 3a: Pedestrian on

Main Road

Figure 3b: Blocked Zebra

Crossing

Figure 3c: Pedestrian

Amenities and Obstructions


7

Table 4: Walkability Survey, Brigade Road

Source: Author


Walking Path Modal

Conflict

Separate pedestrian walkways

High density in peak hours

3

Availability of Walking

Paths

3m-3.2m wide on both sides (see Figure 4a)

Encroached local vendors (see Figure 4b)

Over crowded in peak hours

2

Availability of Crossings Crossing exists at 150m but is not respected

Signalled crossing at 300m

2

Grade Crossing Safety Cars take priority

Difficult to cross, high waiting time (see Figure 4c)

2

Motorist Behavior Honking at pedestrians

Cars take priority, do not respect crossings

2


Street lights at regular intervals

No seats or dustbins

2

Disability Infrastructure Footpath heights are not negotiable by wheelchair

Plenty of obstructions

Excessive crowd for disable person to feel comfortable

0

Obstructions Encroachment from shops

Existing physical barriers like lamp posts

3

Security from Crime Vibrant commercial use

Pick pocketing in the crowd

2

Walking Environment Noise level at 87dB

Air pollution levels high

2

Total 20/50


Footpath

Figure 4b: Vendor

Encroachment

Figure 4c: Pedestrian

Crossing Point


8

5.3 Sampige Road, Malleshwaram

Sampige Road is a commercial street in the

heart of the old neighborhood of

Malleshwaram in Bangalore. The sidewalks

are broad but encroached upon by vendors.

Crossing the street is a problem amidst cars

and two wheelers that do not yield to

pedestrians. Furthermore amenities like

benches or places to rest are missing. Trees

provide comfortable shaded environment to

walk under but also are a major obstruction on

the footpath.

Table 5: Walkability Survey, Sampige Road, Malleshwaram


Walking Path

Modal Conflict

Segregated footpath

Conflict with vendors

3

Availability of

Walking Paths

3.5 m wide on both sides (see Figure 5a)

Encroachment by shops and vendors (see Figure 5b)

3

Availability of

Crossings

No defined crossing except at junctions (100m)

Road markings exist but no pedestrian priority (see Figure 5c)

3

Grade Crossing

Safety

Low as cars have priority and generally do not slow down

Transformers installed at junctions

2

Motorist

Behaviour

Do not yield to pedestrians

Long waiting time

2


Well shaded with trees

Street lights at regular intervals

No seats, public toilets or dust bins

2

Disability

Infrastructure


Plenty of obstructions and walking surface not uniform

0

Obstructions Encroachments from shops and vendors

Existing physical barriers like lamp posts and trees

Obstruction by animals

2

Security from

Crime

Vibrant mixed use

Improvement in lighting is needed

Threat of pick pocketing

2

Walking

Environment

Noise level at 85dB


2

Total 21/50


9

5.4 Gandhi Bazaar Road, Baswangudi

Gandhi Bazaar is a busy commercial street

located in the old neighbourhood of

Baswangudi. Sidewalks are present but

encroached making it impossible to walk on

them. Pedestrians are forced to walk with

motorised traffic. Crossing the street is

difficult as vehicles are fast moving and do not

yield. Furthermore, pedestrian infrastructure

like benches, water points and dustbins are

absent. Cows are also spotted near flower and

vegetable shops and further block smooth

passage. Trees provide shade while walking in

the daytime. However they also encroach most

of the sidewalk.

Table 6: Gandhi Bazaar Road, Baswangudi


Walking Path

Modal Conflict

Pedestrian pathways exist but people end up walking on road due to

encroachment

2

Availability of

Walking Paths

3.2 m wide on both sides

Encroached by 2-wheeler parking, local vendors and shop

encroachments.

Trees and lamp posts also obstruct walking

2

Availability of

Crossings

No defined crossing except at junctions (200m).

People jaywalk to cross

2

Grade Crossing

Safety

Traffic is not controlled hence pedestrian has to negotiate the road

himself

2

Motorist

Behaviour

Pedestrian has to wait for the time when fewer vehicles are on the

road and then cross

2


Street lights

No seats or public toilets

2

Disability

Infrastructure


Plenty of obstructions

0

Obstructions Encroachments from shops 2


Sidewalk

Figure 5b: Vendors on

Footpath

Figure 5a: Pedestrians

Crossing at Junction


10

6.0 ANALYSIS

Tabulating the walkability score for all four

streets reveals that overall Bangalore streets

score low on the GWI. The average score for

the city stands at 20.25/50. On the other hand,

‘Walkscores’ for the same streets generate

high scores (see Table 7). This is on account of

the fact that all streets under study had a

healthy mixed use environment. An average

score of 91.5/100 is indicative of this. This

further illustrates that GWI is a more

appropriate tool is assessing street

environment in the Indian context. Low

walkability scores are primarily attributed to

four factors:

1 Lack of adequate infrastructure or poor

design of the walking environment making

it difficult for people to enjoy their walking

experience.

2 Weak policy and law that does not cater to

pedestrians

3 Weak implementation of footpath

encroachment regulations, traffic

management and pollution standards

4 Lack of education and awareness on part of

pedestrians and motorists in adhering to

basic road rules and regulations.

This analysis advises future attempts at

improving the walking environment by

suggesting that measures must not be restricted

to mere infrastructure improvements. A well

rounded street improvement project must take

into account all these four factors for a larger

and sustained impact on improving

walkability.

Table 7: Walkability Index: Bangalore, India

9th

MainJayanag

ar

BrigadeRo

ad

Sampige

RoadMalleshwar

am

Gandhi

BazaarBaswang

udi

AverageSco

re

Walkability (50)

Modal

Conflict 2 3 3 2 2.5

Path 3 2 3 2 2.5

Existing physical barriers like lamp posts and trees

Security from

Crime

Vibrant mixed use

Improvement in lighting is needed

3

Walking

Environment

Noise level at 80 dB


2

Total 19/50

Figure 6a: Pedestrian on Main

Road

Figure 6b: Vendor

Encroachment Figure 6c: Average Height of

Footpath


11

9th

MainJayanag

ar

BrigadeRo

ad

Sampige

RoadMalleshwar

am

Gandhi

BazaarBaswang

udi

AverageSco

re

Availability

Crossing

Availability 2 2 3 2 2.25

Crossing

Safety 2 2 2 2 2

Motorist

Behaviour 2 2 2 2 2

Amenities 3 2 2 2 2.25

Disability

Infrastructu

re

0 0 0 0 0

Obstruction

s 2 3 2 2 2.25

Security

from Crime 3 2 2 3 2.5

Walking

Environmen

t

2 2 2 2 2

Average City Score 20.25

Walkscore(100)

88 98 90 90 91.5

7.0 CONCLUSION

Recent global developments have shown

that investment in non-motorized modes have

a positive impact on a city’s mobility. Cities

like Amsterdam and Copenhagen are living

proof of such efforts. In this light, there is an

urgent need in Indian cities to take a closer

look at the pedestrian environment and work

towards improving the safety and comfort

standards for this. In this context GWI is a

comprehensive tool that can be adopted to rate

streets, especially in the Indian context.

Adding the 10th criteria for ‘Environment

Quality’ can further help elaborate this

method. Availability in the form of a smart

phone app is a great initiative that makes it

easy to use and generate street scores. It can

further encourage citizens to engage in

reporting scores for their local streets. Also the

availability of scores on an online open source

platform will help researchers access and use

this information. Analyzingresults, based on

the cause of low scores, aims at assisting

municipalities to device better and well

rounded policies and projects to improve

walkability.

REFERENCES

1 TERI. (2010). Air Quality Assessment,

Emission Inventory and Aource

Apportionment Study for Bangalore City.

TERI, New Delhi.

2 Ministry of Urban Development. (2008).

Traffic & Transportation Policies and

Strategies in Urban Areas in India.

Ministry of Urban Development, India

3 Indian Institute of Human Settlements.

(2011). Urban India 2011: Evidence.

Autumn Worldwide.

4 Abley, S. (2005).Walkability Scoping

Paper. http://

www.levelofservice.com/walkability-

research.pdf [accessed 15 April, 2014]


12

5 Krembeck, H. & Shah, J. (2006). The

Global Walkability Index. Massachasetts

Institute of Technology, MIT Libraries

6 Leather, J. (2011).Walkability and

Pedestrian Facilities in Asian Cities. Asian

Development Bank, Philppines\

7 RITES India. (2011). Comprehensive

Traffic and Transportation Plan for

Bengaluru. Karnataka Urban

Infrastructure Development and Finance

Corporation.

URBAN TRANSPORT

JOURNAL


* PG Student, Civil Engineering Department, The M.S.University of Baroda, [email protected]

**Associate Professor, Civil Engineering Department, The M.S.University of Baroda, [email protected]

# PhD Scholar, Civil Engineering Department, SardarVallabhbhai National Institute of Technology, Surat,

[email protected]

##Assistant Professor, Civil Engineering Department, The M.S.University of Baroda,

[email protected]

EVENT DAY EFFECT ON PEDESTRIAN CHARACTERISTICS FOR

CBD STREET OF INDIAN METROPOLITAN CITY

Hardik S Sukhadia*, Sanjay M Dave**, Jiten Shah#, Dipak Rathva

##

Abstract: Walking is one of the most important, economical, flexible and eco-friendly mode of transportation.

In India, walking on carriageway instead of sidewalk is widespread scenario reasons being sidewalks are

inadequate, encroached by hawkers, vendors and unprotected with guard rail. The situation becomes worst on

CBD streets during event occasion and weekends peak hours. This study is aimed on pedestrian flow behaviour

on normal week day and event day. The study was carried out by video recording of pedestrian movement in

CBD area of Vadodara city for one normal and two event days. The analysis helps in accessing the behaviour,

volume, speed, density, space and level of service of pedestrian flow in both conditions. Pedestrian flow on

normal working days was laminar one with somewhat queuing pattern observed for most of the survey period

with average rate of flow 10 pedestrian/min and maximum rate of flow 22 pedestrian/min during peak hours. On

other hand platoon effect was observed for religious event day of Navratri Durgashtami (Event day-1) with

average rate of flow 22 Pedestrian/min and maximum rate of flow 45 Pedestrian/min during peak hours. The

average pedestrian speed calculation worked out for study stretch was observed 0.984 m/sec for normal working

day and for the religious event day observed 0.903 m/sec. The comparative pedestrian characteristics for two

conditions revealed that some TSM action to improve pedestrian movement is necessary on event day.

Keywords: Pedestrian Characteristic, walking speed, CBD, Event day, Normal working day

1.0 INTRODUCTION

Waking is most reliable, sustainable, health

gainer activity. Also the walking is one of the

cheapest modes of transportation with highly

contributing safe and liveable environment.

Every journey starts and ends with walking

trips. In developing country at least 40%

journeys up to 1-2 km are walking trip.

(Source: ADB Bank and HCM 2000). In such

countries different areas of the metropolitan

and 2nd

order metropolitan cities share large

number ofwalking trips. Among these areas

CBD areas generates majority walking trips.

Sidewalks have been placed both sides of the

carriageway for easy and safe movement of

pedestrian mobility without interacting to

vehicle traffic. In developing country like

India the vehicle growth rate ishigh in last

decade. Also urbanization rate of metropolitan

cities is high in last decades. This resulted in

more importance to planning of facilities for

vehicular movement and negligence towards

pedestrian environment. In order to achieve

high speed and rapid connectivity for vehicular

traffic, there is stress for the expansion of

carriageway width. In typical CBD areas of

historical cities of India, where old planned

city roads suffers from lack of space is main

problem. This trend leads the curtailment of

the sidewalk. In addition, the problem

encroachment due to hawkers and vendors

reduce the effective walkway width of the

sidewalk. This compels to pedestrian walk on

the carriageway. The situation becomes very

chaotic to control on event day such as

religious festival days. On event day in Indian

condition hawkers and vendors encroach near

to the religious place and also the flow of

pedestrian movement increases. Sometimes

flow of pedestrian shifts to the carriageway.

The pedestrian flow on the carriageway creates

friction with vehicular traffic. Due to this

interaction between vehicle and pedestrian

there are chances of accident or the problem

ends with the traffic jam. This problem of

Event Day Effect on Pedestrian Characteristics for CBD Street of Indian Metropolitan City

14

pedestrian requires some Transport System

Management (TSM) action because such event

based pedestrian flow occurs very frequently

in India. Also in India, very little attention has

been paid to study pedestrian behaviour;

particularly in CBD area hence this study will

provide some base for investigating pedestrian

behaviour under influence of religious and

festival event and also help local authority to

frame appropriate measures for safe and

comfortable movement of pedestrian.

2.0 LITERATURE REVIEW

Very few research studies have been

reported on pedestrian characteristics in India.

Literature on pedestrian studies is quite diverse

to the vehicular traffic. The pedestrian studies

conducted abroad include pedestrian flow

characteristics and modelling of flow

parameters. Many researchers have examined

the influencing factors of pedestrian

characteristics. Some of the researchers Fruin

(1971), Polus et al. (1983), Tarawnch (2001),

Montufar et al. (2007), and Finnis and Walton

(2008) observed that walking speed of female

pedestrian is slower than male pedestrian and

walking speed reduces with age of pedestrian.

The speed of pedestrian is also affected by

type of pedestrian (i.e. local or outsider)

Tanabariboon et al. (1986) observed that the

mean walking speed of the Singaporeans is

slower (74m/min) in comparison with speed of

U.S. pedestrians (89m/min). Polus et al. (1983)

found that the average walking speed for

pedestrian of Israel is 79 m/min, while

Khoushki (1988) found 65 m/min in Riyadh

and Morrall et al. (1911) found the walking

speed 75 m/min in Colombo, Sri Lanka

respectively. RajatRastogi (2010) found the

average walking speed 72 m/min for Indian

pedestrian.

Tarawnch (2001), Carey (2005), and K.

Singh and P.K.Jain (2011) found that group

size affect the walking speed significantly.

P.K.Jain (2011) concluded that effect of group

size on walking speed is low for group size up

to 3 and high for group of five or more. Carey

(2005), Montufar et al. (2007), K.K.Finnis and

Walton (2007), and Jianhong (2012) found

that younger pedestrian are faster than older

and children.

Dammen and Hoggendom (2005) observed

that pedestrian walking speed depends on

walkway characteristics such as width, type of

facility (i.e. with or without guardrail) and

environmental factor. The walking speed of

pedestrian is also affected due to activity

performed during walking. Morrall et al.

(1991), K.K. Finnis and Walton (2007),

Ronald Galizo and Luis Ferriro (2012),

Jianhongchen and Nanjing jian (2012), Kotkar

Kishor et al. (2010) found that pedestrian

walking with luggage are slower than those

pedestrian who has no luggage. Young (1998)

found that walking speed of the pedestrian is

significantly differing from those wearing

headphones and talking on cell phones. The

land use pattern is also been subject of

research. Al-masuied et al. (1993), K. Singh

and P.K. Jain (2011) found that the walking

speed is differing with different land use such

business area, residential area, educational

area. They also found that surrounding

environment is an important factor which

affects the walking speed of pedestrian. Lam

and Chang (2000) observed that pedestrians

walking in commercial areas are faster than

those in residential areas. Finnis and Walton

(2007) found the walking speed of pedestrian

in indoor walkways is slower than outdoor

walkways.

K.K. Finnis and Walton (2007) observed

that commuters have significantly higher

walking speed than others. He also observed

that pedestrian talking to other and observing

surrounds has slower speed than commuters.

He also found that the pedestrian who wears

flip-flop shoes are slower than others. He also

found that there was no effect of gradient on

walking speed up to 0o

to 4o. However, the

walking speed significantly reduces at gradient

more than 5o.


15

Al-Masaeid et al. (1993) developed the

pedestrian speed flow relationship for the

central business district (CBD) areas in Irbid,

Jordan. Quadratic polynomial regression

relation was found to be the best fit. They

observed that in the design of CBD sidewalks,

pedestrian demand to capacity ratio should be

limited to 0.5. Pedestrian flow was analyzed

on the basis of effective width of sidewalk

rather than the lane concept.

Many researchers have studied pedestrian

flows accross the country as well as abroad

under different conditions like outdoor

walkway, sidewalks in central business district

(CBD) areas, under mixed traffic conditions,

prevailing pedestrian flow unidirectionally and

bidirectionally. They have devoloped flow

speed relationship. Tanaboriboon et al. (1986)

developed flow relationships for sidewalks and

walkways in Singapore and compared them

with those obtained for the United States and

the Britain.The relation between speed and

density becomes exponential under heavy

pedestrian flow. Polus et al. (1983) developed

single and three regime linear speed-density

models for pedestrian flows on sidewalks in

CBD of Haifa (Israel).In similar type study Al-

Masaeid et al. (1993) found that the quadratic

polynomial relation fits the speed-flow data

the best relationship between speed and

density was found to be linear while flow-

density and flow-speed relationships were

quadratic.

3.0 OBJECTIVES AND SCOPE OF

THE STUDY

1. To study the pedestrian movement

characteristics on normal working and

event day.

2. To evaluate Level of service for the study

area.

3. To recommend the appropriate TSM action

for smooth movement of pedestrian due to

event effect in study area.

4. The scope of this study is limited to the

typical CBD street attracting considerable

pedestrian trips on normal working day and

substantial rise of the same on event days.

4.0 STUDY AREA

Vadodara is third largest city of Gujarat

and eighteenth largest city of India. It is

situated on banks of river Vishwamitri. It has

an area of 148.95 km2 and urban population of

1.8 million (Census 2011). It is known as

‘cultural city’ and it has reputation as

educational hub as well as chemical hub of

Gujarat. Vadodara has 1680 km paved and 400

km of unpaved roads.Average trip length for

city is about 6 km, due to which mode share is

inclined towards private vehicles comprising

about 50% of total modal split. Public

transport is one of the areas, which has been

lacking far behind in the city. Inadequacy of

public transportation system restricts its mode

share to 6% while 15% of trips are taken care

by IPT (autos) and 29% by non-motorised

transportation (NMT). The percentage share of

walking trip is found to be 66% of total NMT,

particularly in CBD areas (RITES 2006).

The selected study stretch is located in one

of the major arterial corridor of the oldest

CBD area covered with four gates called as a

walled city (Figure 3). The study stretch was

selected in such a way that it comprises mix

land use like commercial activity at ground

level and residential at first floor level and

above as well as having good number of

religious places resembling to typical Indian

cities. The main attraction toward wall city is

historical temples, jewellery shops and cloth

market, which generate huge amount of

pedestrian movement. The sidewalk is only the

facility, which provides pedestrian movement

and is generally encroached by hawkers and

vendors. This is common features of the study

stretch and this activity increases during event

day resulting in reduction of effective

walkway width. Due to reduction in space,

pedestrian have restricted choice for

movement and to gain walkable speed they

share the main carriageway of vehicular flow.

This hinders vehicular traffic and highly


16

affects their safety too. Especially in event day

pedestrian movement is quite high compared

to normal working day which requires to be

addressed considering safety and smooth flow

of pedestrian. This study mainly focuses on the

comparison of pedestrian movement on

normal working day and event day.

5.0 DATA COLLECTION AND

METHODOLOGY

The study stretch included typical CBD

Street having high commercial activities and

religious temples. A study was carried out for

three different days and locations on the same

street. These included a normal working day

and two event days of Durgashtami (Event

day-1) and Dushera (Event day-2) of Navratri

which is predominant festival. The selection of

study stretch has been done in such a way that

it includes commercial and residential land use

on both sides of the road, freedom from

encroachment by hawkers and vendors and at

least 15 meter away from cross road to have

fair assessment of pedestrian characteristics.

The detailed cross section of study location is

shown in Figure 1. Video-graphic technique

was employed for collecting the pedestrian

data. A strip of known length (6 meter) was

marked on the sidewalk by using a white or

yellow oil paint depending on floor material

for measurement of walking speed and flow of

the pedestrian. The pictorial view of study

stretch with trap marking is shown in Figure 2.

A high mega pixel (14.0 MP) video camera

was used for accurate data collection, which

was installed at an elevated point so that it is

possible to covers the pedestrian movement on

the entire strip of sidewalk as well as cover the

half of the carriageway for the pedestrian flow

on carriageway. The pedestrian movement was

recorded during 10:00 to 17:00 hrs for the

normal and event days. The required

pedestrian data were later extracted from the

Figure 1: Cross Section of the Study Stretch on M.G. Road

Figure 2: Strip Marked on Sidewalk


17

recorded videos by using video player

operating at normal speed. The basic

parameters of pedestrian flow data were

extracted on one minute basis through

observing entry and exit movement of

pedestrians in the marked strip length. The

characteristics of surrounding area were such

that pedestrian could not walk on the dedicated

path due to encroachment by hawkers and

shopkeepers to display their product, resulting

in pedestrian spill over on the main

carriageway. Pedestrian observation was

carried out on sidewalk and carriageway

considering directional movement, gender,

age, and carrying of luggage/children.

For pedestrian flow, the number of

pedestrian passing the marked line was noted

per minute basis. The time taken by pedestrian

to cross the strip length was noted with least

count of 0.01s stopwatch to determine

pedestrian speed. At least five pedestrian of

each category (age, gender and luggage or

without luggage) were selected for speed

analysis per minute.

Samples for the analysis of data were

collected from total 420 minute of

videographic survey of each day. Based on the

pedestrian flow, pedestrian density per square

meter was calculated by pausing video at

interval of every 3 second.

Figure 3: Study area location

6.0 RESULT AND ANALYSIS

The classified data of 38,108pedestrians

was extracted from video considering

movement on sidewalk and carriage way i.e.

closed to parking. It was observed that

composition of pedestrian on sidewalk and

carriageway is dominated by female pedestrian

with about 65% share of total pedestrians on

normal working day as well as on event day.

From the result it is observed that movement

of pedestrian is quite high in event days due to

the peculiarity of the CBD area such as major

whole sale and retail cloths, jewellery markets

as well as some religious place (Figure 4). It is

also observed that about 20% and 33% of total

volume of pedestrian walking on carriageway

on normal working and event day respectively

(Figure 5)

Figure 4: Pedestrian Movement on Sidewalk


18

Figure 5 shows significant movement of

pedestrian on carriage way and this is due to

encroachment and other activities on sidewalk,

resulting in reduction in effective walkway

width. Also those pedestrian who want

continuous and uninterrupted walk are

unwillingly forced to share the carriage way

and their volume reaches to 40% to 50% of the

total volume, particularly on event day. Such

trends shows pedestrian put their life and other

vehicular flow in risk. It indicates that there is

an urgent need to improve the pedestrian

environment on the sidewalk for better

performance of pedestrian activities as well as

smooth vehicular flow.

The study also examines the composition of

pedestrian characteristics based on age,

gender, luggage carrying pedestrian and

directional movement on sidewalk. From the

data it is found that during normal working

day from total (4,567) pedestrian movement in

downstream (2,462) is higher than the

upstream (2,105), during event day-1 from

total (8,433) pedestrian movement in upstream

(4,640) is higher than the downstream (3,793)

and during event day-2 from total (10,075)

pedestrian movement in downstream (52,42) is

higher than the upstream (4833).

Figure 6 shows the gender wise

composition of pedestrian movement in both

Figure 5a: Pedestrian volume on half hourly basis for sidewalk and carriageway

Figure 5b: Pedestrian volume on half hourly basis for sidewalk and carriageway

Figure 5c: Pedestrian volume on half hourly basis for sidewalk and carriageway


19

the direction. Figure 6 illustrate that the

ascending pedestrian flow constitutes 34%

male and 66% female where as in downstream

direction 35% male and 65% female are

observed. The composition of female

pedestrian was dominated by about 65% of

total pedestrian volume on normal working as

well as on event day. Pedestrian were grouped

into three categories as children (< 15 years),

young (15-50 years) and elder (> 50 years).

The proportion of children, younger and elder

is 10%, 84% and 6% respectively in both the

directions.

Figure 6: Gender wise pedestrian composition for normal day

Figure 6c: Gender wise pedestrian composition for event day-2

Figure 6b: Gender wise pedestrian composition for event day-1


20

It was also found 2-4% of total pedestrian

was found walking with luggage in both the

directions. The walking speed is subjected to

pedestrian characteristics such as gender, age,

walking with luggage and direction of

movement. From the analysis of speeds of all

selected pedestrians, the average walking

speed on sidewalk was 1.0 m/s on normal

working day and 0.91 m/s on event-1 and 0.9

m/s on event-2. This reduction in speed of

pedestrian on event day may be accounted for

their wandering behaviour on display of

products and attractive offers by the

shopkeepers who encroaches the side walk

also. It has also a side effect on the pedestrian

who follow them. Table 1 provide the

statistical analysis of mean walking speed for

various conditions.

Table 1a: Statistical analysis of mean walking speed for normal working day

Facility Gender Loading

Condition

Sample

size

Mean

(m/s) Median S. D Variance CV

Up

Str

eam

Male W/O Luggage 1,355 1.146 1.165 0.069 0.005 0.059

With Luggage 40 1.12 1.163 0.108 0.012 0.094

Female W/O Luggage 2,332 1.183 1.149 0.060 0.004 0.052

With Luggage 79 0.902 1.158 0.087 0.007 0.076

Dow

n S

trea

m

Male W/O Luggage 1,162 1.159 1.179 0.064 0.004 0.054

With Luggage 31 1.123 1.231 0.136 0.019 0.116

Female

W/O Luggage 1,641 1.184 1.159 0.066 0.004 0.057

With Luggage 62 1.106 1.135 0.101 0.010 0.088

AGE

Sid

ewalk

Children --- 595 1.065 1.056 0.066 0.004 0.062

Young --- 5,373 1.213 1.210 0.037 0.001 0.031

Elder --- 734 0.707 0.721 0.120 0.014 0.170

Table 1b: Statistical analysis of mean walking speed for Event-1 day (Durgashtami)


Condition

Sample

size

Mean

(m/s) Median S. D Variance CV

Up

stre

am

Male W/O Luggage 1,326 1.177 1.137 0.178 0.032 0.151

With Luggage 52 1.111 1.125 0.146 0.021 0.131


With Luggage 128 0.898 0.925 0.107 0.011 0.119

Do

wn

stre

am

Male W/O Luggage 1,518 1.189 1.164 0.161 0.026 0.135

With Luggage 73 1.099 1.045 0.368 0.136 0.335

Female

W/O Luggage 2,786 1.044 1.036 0.136 0.019 0.131

With Luggage 185 0.933 0.942 0.176 0.031 0.189

AGE

Sid

ewalk

Children --- 518 1.052 1.053 0.155 0.024 0.147

Young --- 7,334 1.131 1.125 0.140 0.020 0.124

Elder --- 515 0.559 0.539 0.143 0.020 0.255


21

Table 1c: Statistical analysis of mean walking speed for Event day -2 (Dushera)


Condition

Sample

size Mean Median S. D Variance CV

Up

Str

eam

Male W/O Luggage 1,349 1.144 1.192 0.099 0.010 0.083

With Luggage 89 1.126 1.131 0.082 0.007 0.072


With Luggage 201 1.120 1.128 0.086 0.007 0.075

Do

wn

Str

eam

Male W/O Luggage 1,474 1.143 1.188 0.128 0.016 0.108

With Luggage 77 1.122 1.164 0.076 0.006 0.065


With Luggage 233 1.182 1.184 0.099 0.010 0.085

AGE

Sid

ewa

lk Children --- 1,437 1.105 1.103 0.068 0.005 0.062

Young --- 8,032 1.224 1.226 0.055 0.003 0.045

Elder --- 272 0.564 0.535 0.099 0.010 0.176

Table 1 (a, b, c) shows that in both, normal

as well as event days, walking speed on

upstream is found lower (1.01m/s) than the

downstream (1.2m/s). The results reveals that

average walking speed is quite lower

compared to the speeds reported in literature

RajatRastogi et.al (2010). The mean speed of

pedestrian carrying luggage for all the

categories: upstream and downstream in both

normal as well as event day was lower than the

without luggage carrying pedestrian except

event2, female walking in downstream. The

deviation in mean speed was higher in

pedestrian with luggage on normal working

day. This might be due to small sample size

having affected by higher percentage of

without luggage carrying pedestrian that

means they must follow the speed of stream

flow contributing higher percentage of

pedestrian without luggage. For that reason

there was close difference in average walking

speed of without and with luggage pedestrian

except for females moving in upstream

direction. Whereas in event days, deviation

and coefficient of variance of pedestrian speed

is lower in luggage carrying condition with

exception of event day 1 downstream

pedestrian flow. This may be due to volume of

pedestrian moving with luggage was quite

higher than the normal day and luggage

condition may restrict their speed to some

extent. Table 1(a, b, c) proves that there is

significant effect of luggage on the walking

speed. The higher walking sped is observed by

younger pedestrian followed by children and

elder. The interesting finding for all cases is;

average waling speed of elder pedestrian is

almost half of the children.

On normal working day, pedestrian

movement gradually increases up to evening

with peak observed evening as shown in

Figure 5. Whereas pedestrian rush found to be

constant throughout the day time with one

peak in afternoon in both event days. The

same variation is reflected in term of density

as shown in Figure 7. As volume increases

density of pedestrian also increases.


22

Figure 7 shows increase in density with

time and generally density represent the

congestion level. As the density increases, it is

difficult to find the space for the further

movement of pedestrian and it restricts the

speed. In both event days, large pedestrian

movement was found with relatively higher

percentage of luggage carrying pedestrian

which reduced the service level of pedestrian.

The level of service (LOS) describes the

comfort level of pedestrian. It is observed that

average pedestrian density is three times on

event day-1 and event day-2 (0.3 ped/m2) than

the pedestrian density on normal working day

(0.1 ped/m2), which reflects deterioration in

the LOS for the pedestrian movement. Table 2

and Table 3 shows the current Level of Service

for existing facility of study area.

Figure 7a: Time series plot for Density variation (Normal day)

Figure 7b: Time series plot for Density variation (Event day-1)

Figure 7c: Time series plot for Density variation (Event day-2)


23

Table 2: Volume based LOS for existing facility of study area

From Table 2 it is seen that the level of

service evaluated on basis of volume

concentrates is around C for normal working

day and it declines up to D for event day. It

indicates that some measures are required for

the smooth pedestrian movement by some

TSM action like prohibition of on street

parking, hawkers and roadside vendors

particularly on event days. Pedestrian space is

characterized as reciprocal of density.

Considering area module and volume, the LOS

for normal working day was observed to be D

but on event days it declined up to E

considering the area module as shown in Table

3 and which reveals that the sidewalk

functions with lower level of performance.

This may be due to the higher percentage of

the luggage carrying pedestrian. Noteworthy

that the study results does not included the

movement of pedestrian who were walking on

carriage way on event and normal days. If

included, further declination of LOS up to E or

F may have been observed. The study results

give an idea about the pedestrian flow

characteristics on both normal as well as

various event days and pointed out the need of

TSM action for better performance of side

walk.

Table 3: Comparative Level of service based on space (m2/p)

Flow rate Normal

working day

Level of

Service

Event days Average of

Event day-1

and 2

Level of

Service Day-1 Day-2

Maximum 3.70 D 1.079 1.29 1.18 E

Average 7.092 B 2.19 2.46 2.46 D

7.0 CONCLUSION

The study carried out in CBD area of

Vadodara city revealed that pedestrian have to

endanger their safety for mobility on the event

day. From the study, it was found that the

pedestrian volume in study area increased up

to 90% on the event day as compare to normal

day. The significant bi-directional movement

observed in study area induces friction and

restricts comfortable movement within the

sidewalk. As compared to normal working day

the mean walking speed decreased up to 10%,

density increased 80 to 100% and space

decreased 70 to 100% on event days. The

study also found that almost 40 to 50% of total

pedestrian walk on carriageway which may

cause obstruction to the smooth vehicular

traffic flow.

The comparative study regarding pedestrian

flow, speed, density and space revealed that

pedestrian environment degrades to significant

extent on the event days. Considering the fair

number of religious event and festival

occurring during a year, a permanent strategy

to improve the safety and smooth flow of

pedestrian is necessary in Indian context. It is

proposed to prohibit on street parking on event

days and implement effective enforcement

Pedestrian

Flow

In

Normal

working

day

(p/m/min)

LOS As

per

(HCM

2010)

In Event day Avg. flow

on event

days

(p/m/min)

LOS

As per

(HCM

2010)

% Change

Normal

and Event

day

Day-1

(p/m/min)

Day-2

(p/m/min)

Maximum 22 D 26 36 31 D 41%

Average 16 C 23 25 24 D 50%


24

measures to prevent the encroachment of

sidewalk. This TSM action may be improve

the pedestrian environment in CBD area.

REFERENCE

1. Al-Masaeid, H.R., Al-Suleiman, T.I. and

Nelson, D.C., (1993). Pedestrian Speed

Flow Relationship for Central Business

Areas in Developing Countries.

Transportation Research Record 1396,

National Research Council, Washington,

69-74.

2. Dammen, W., and Hoogendoorn, S. P.

(2003). Experimental research of

pedestrian walking behaviour, Proceedings

of Annual Meeting of Transportation

Research Board, CD-ROM, National

Academy Press, Washington.Guidelines for

pedestrian, Indian Road Congress-103,

1988.

3. Finnis K.K., Walton D., (2007). Field

observations of factors influencing

pedestrian walking speed.

4. Galiza R., Ferreira L., (2012). A

methodology for determining equivalent

factors in heterogeneous pedestrian flows.

5. Highway Capacity Mannual (2010).

6. Jianhong Y., Xiaohong C., Nanjing J.,

(2012). Impact analysis of human factors

on pedestrian traffic characteristics.

ELSVIER, The fire safety journal.

7. Kotkar K. L., Rastogi R. and Chandra S.

(2010). Pedestrian Flow Characteristics in

mixed flow condition, ASCE, Journal of

Urban Planning & Development,136(3)22-

23

8. Koushki, P. A. (1988). Walking

characteristics in Central Riyadh, Saudi

Arabia. Journal of Transportation

Engineering, 114(6), 735–744.

9. Larusdottir A. R. and Dederichs A. S.,

(2010). Evacuation of children: movement

on stairs and on horizontal plane. Journal

of Fire Technology.

10. Leather J., Fabian H., Gota S., and Mejia

A., (2011). Walkability and pedestrian

facilities in Asian cities. ADB Sustainable

Development Working Paper Series.

11. Montufar M., Arango, J., Porter, M., and

Nakagawa, S. (2007). Pedestrians ‘normal

walking speed and speed when crossing a

street. Transportation Research Record,

Transportation Research Board,

Washington, DC, 90–97.

12. Parviz A.K., (1988). Walking

characteristics in central Riyadh, Saudi

Arabia. ASCE, Jour-nal of Transportation

Engineering.

13. Polus, A., Schofer, J. L. and Ushpiz, A.

(1983). Pedestrian flow and level of

service, ASCE, Journal of Transportation

Engineering, 109(1), 46-56.

14. Rastogi R., Thaniarasu I., Chandra S.,

(2011). Design implications of walking

speed for pedestrianfacilities. ASCE,

Journal of transportation engineering.

15. Singh K., Jain P.K, (2011). Methods of

assessing pedestrian level of service.

Journal of Engineering Research and

Studies.

16. Tanaboriboon Y., Hwa S. S, and Chor C.

H., (1985) Pedestrian characteristics study

In Singapore. ASCE, Journal of

Transportation Engg.

17. Young, S. B. (1998). Evaluation of

pedestrian walking speeds in airport

terminals. Transportation Research Record

1674, Transportation Research Board,

Washington, DC, 20–26.

URBAN TRANSPORT

JOURNAL


* Post Graduate Student, Transportation Engineering Group, Dept. of Civil Engineering, Indian Institute of

Technology (IIT) Roorkee, [email protected]

** Associate Professor, Transportation Engineering Group, Dept. of Civil Engineering, IIT Roorkee,

[email protected]

EFFECT OF LANE FRICTION ON SPEED OF NON-MOTORIZED

VEHICLES

Prasham Khadaiya* and Rajat Rastogi**

Abstract: Traffic flow on roads consists of three components, i.e. motorized vehicle flows, non-motorized

vehicle (NMV) flows and pedestrian flow. This paper concentrates on the NMV (bicycle and cycle rickshaw

together) flows on urban roads. The study was carried out in Roorkee city, a small sized city with high intercity

and local traffic on its roads. Data were collected through a combination of manual and videography methods on

three roads, all of two-lanes with either one-directional or two-direction traffic. The impacts of roadway space

and flow constraints and traffic composition and flow on speed of NMVs were studied. The analysis indicated

that on a road segment with no friction from opposite direction, the NMV speed is reduced with an increase in

total flow; on segment with friction from opposite direction it is reduced with increase in opposing direction

flows, while on segment with all frictions it is reduced with NMV, total and opposing direction flows. The gap

between speed ranges of NMVs and MVs also reduced under the three scenarios. The percent share of NMVs in

total flow also influenced the speed profiles.

Keywords: Non-motorized Vehicles, Speed, Lane Frictions, Composition, Flows

1.0 INTRODUCTION

The non-motorized vehicles (NMVs) are

those vehicles which require human power for

their mobility. NMVs offer low cost transport,

are environment friendly (no pollution), use

renewable energy, and emphasize on the use of

labour rather than capital for mobility. These

are well suited for short trips in most cities

regardless of income and offer an alternative

to motorized transport especially under traffic

congestion conditions. These can be

considered as an appropriate element in

strategies dealing with poverty alleviation, air

pollution, management of traffic problems and

motorization, and the social and economic

dimensions of structural adjustment. These

play a complementary role to public

transportation. Varieties of non-motorized

users are present on roads in different

countries and their share is reported to be

continually increasing over years. Each of the

vehicle type has different flow characteristics.

With the increasing variety of emerging NMV

users comes the question whether we are

designing and building suitable facilities for

them? Many jurisdictions throughout the

United States have adopted the American

Association of State Highway and

Transportation Officials (AASHTO, 1999)

Guide for the design, layout and development

of bicycle facilities. In India, IRC 11-1962

outlines the recommended practice for the

design and layout of cycle tracks. This guide is

written with bicyclists in mind and is based on

old research. This paper focuses on the

possibilities of contributing to the existing

code of practice and incorporates the bicyclists

and cycle rickshaws for their study. In the

process, the impact of flow composition,

operational characteristics of roads and flow

and space frictions on speed of cycle, cycle

rickshaw and both combined as NMVs is

examined and studied.

Not much work is reported in literature

regarding the flow analysis of NMVs in

conditions similar to that in India. Navin

(1994) experimentally determined the

operating performance of a single bicycle and

traditional traffic flow characteristics of a

stream of bicycles. These were compared with

the observed data. Wang and Wu (2003)

examined the characteristics of motorized,


26

non-motorized and pedestrians individually

and as well as the interference among them in

mixed traffic conditions. Duthie et al. (2010)

examined the impact of design element

including the type and width of the bicycle

facility, the presence of adjacent motor vehicle

traffic, parking turnover rate, land use and the

type of motorist-bicyclist interaction. Hongwei

et al. (2011) found that when a non-motorized

lane is provided adjacent to the curb parking

space, the effective lane width of the non-

motorized lane decreases. El-Geneidy et al.

(2013) examined travel speed of bicyclists on

various types of facilities, namely on-street,

off-street, and mixed traffic. In India IRC: 11-

1962 provide the guide lines for cycle tracks

such as need of cycle track, capacity,

horizontal curves, vertical curves, sight

distance, lane width, vertical clearance,

horizontal clearance, etc.

2.0 DATA COLLECTION

2.1 Study Area

Roorkee was selected as the study area to

conduct the study related to NMV flows. The

city has a semi urban-rural setting with lots of

daily trips being made to the city from

adjoining villages. The city is one of the entry

points to the Uttarakhand state and caters to

the traffic coming from Muzaffarnagar side

and going to Haridwar, Dehradun and beyond.

It is on NH-58 and NH-73. The city has high

proportion of NMVs in the total flow.

2.2 Location Characteristics

To select a suitable location for carrying

out NMV study, following general points were

given consideration:

NMVs share in the total traffic is high

Effective width of the road remains

uniform along the segment considered

No intersection falls within the segment

and adjacent it which can influence the

speeds

The road segment is clearly visible from an

accessible vantage point

Apart from the above considerations,

following flow considerations were also

considered:

Number of lanes were fixed to two with

adequate shoulders on sides

The possibility of segregation of traffic by

way of provision of a median

Possible variation in the flow and its

composition

Considering the above points, three data

collection locations were identified. These

were:

a) Segment between MalviyaChowkand BSM

Y-intersection on NH-73

b) Civil lines Road near petrol pump

intersection

c) Rampur Road (Old NH road)

These locations are referred as Location-1,

Location-2, and Location-3 respectively in rest

of the paper. The physical features of the road

segments are given in Table 1.


27

Table 1: Physical features of the selected road segments

Data

collection

location

Number

of lanes

Width of

carriageway

(m)

Roadway

width (m)

Flow direction Friction level

Location 1 2 7.00* 8.20* One-directional Median on one side,

negligible influence of

parking of vehicles

Location 2 2 8.00 11.00 Two-directional Opposing traffic,

higher influence of

parking of vehicles

Location 3 2 6.20 10.70 Two-directional Opposing traffic, high

influence of parking of

vehicles and shops on

sides

This is half of the divided carriageway / roadway. The road is 4-lane divided.

The level of friction as mentioned in the

table indicates that the effective width

available to the vehicles is lower than what is

given. Location 2 and 3 have some open space

in front of the shops but that was consumed

mainly by the shops. Therefore, the level of

friction has increased as we move from

location 1 to location 3.The three locations are

shown in Figure 1.

a) Location 1 b) Location 2 c) Location 3

2.3 Method of Data Collection

Video graphic method was used to capture

the traffic flow at the selected locations. The

camera was fixed at an elevated position,

usually at the top of an adjoining building, to

obtain an overall view of the selected test

location. The data were collected in the

morning and evening hours when the flow was

high. The physical features of the test location

were measured using a 30 m tape. Speed of the

vehicle was estimated with respect to a trap

length being marked on the carriageway. The

details of the data collection effort at different

locations are given in Table 2.

Figure 1: Vehicular movement at different locations


28

Table 2: Details of data collection at different locations

Data collection

location

Time period Duration (minutes) Length of segment (m)

Location 1 8:30 am to 10:40 am 130 16.0

Location 2 10.30 am to 12.00 am

4.00 pm to 5.30 pm

180 24.0

Location 3 10.00 am to 11.00 am

5.00 pm to 6.00 pm

120 10.0

2.4 Data Processing

Data were processed in the office by

playing video graph on the monitor. The flow

characteristics of the cycle, cycle rickshaw and

other motorized vehicles were measured

separately considering a two minutes interval

using system stop watch. Data were recorded

in the MS-Excel work sheet for further

processing. Based on the requirements the

time intervals of successive two minutes were

converted into four minutes interval. Flow of

motorized and non-motorized vehicles was

extracted in every 2-min time interval and this

was converted to equivalent hourly flow units

by using appropriate multiplier. Two white

lines were marked on the carriage way at

distances as mentioned in Table 2. The time

taken by a vehicle in traversing this distance

was noted using system stop watch. Thus, the

speeds of the vehicles were estimated

3.0 NMV FLOW-

CHARACTERISTICS

2.5 Traffic Composition

The traffic flow composition at different

locations is given in Figure 2.

The specific nature of the road section can

be studied from the traffic composition as

shown above. First location shows almost all

categories of the traffic which is true as this

falls along NH-73. Second location is

shopping area with open spaces and restriction

on movement of heavy vehicles and

commercial vehicles. Cars are also allowed for

lesser time periods. This is clear from only 3%

share of cars and nil presence of heavy and

commercial vehicles. The third location is in

old city which is densely occupied. This

location has shops on both the sides which

occupy spaces on the road side thus pushing

the vehicles to be parked to the shoulders.

Reduction in effective space has resulted in

absence of big size vehicles at this location.

The share of NMVs on the three locations is

33%, 46% and 51% respectively. Within the

NMVs, the share of cycle rickshaw is 18% at

first location and 41% at rest of the two

locations. The location 2 and 3 are shopping

areas and hence is the reason of increase in the

share of cycle rickshaws.

(a) Location 1 (b) Location b (c) Location 3

Figure 2: Composition of traffic at three locations


29

2.6 Space Occupied at Road Side

Observations were taken regarding the

space being occupied by NMVs and bicycle

and cycle rickshaw individually from the edge

of the carriageway. This would provide an

initial account of the space that need to be

marked on roads for the exclusive usage of

NMV modes. Figure 3 presents the overall

space occupied by the NMV category at the

three locations. It can be observed that

majority of NMVs are moving within 2m from

the edge of the carriageway on location 1 and

on one side of the location 2. Side friction in

terms of parking of vehicles has caused low

usage of space within 1 m from the edge of the

carriageway. Parking of bigger vehicles has

shifted the NMV flow towards the centre of

the carriageway. Location 3 plot clearly show

the impact of space being occupied by the

shops and the parking of vehicles on road side

in terms of the higher usage of middle space

on the carriageway. This is between 2 to 3 m

on side and from 1 to 4 m on the other side of

the carriageway.

Similar analysis was carried out for the

constituents of NMV flow i.e. bicycle and

cycle rickshaws. The trend was found to be

same at all the locations. In the case of higher

side frictions the cycle rickshaws were found

moving keeping a buffer at the side of the

carriageway.

2.7 MV and NMV Speed Variation

The speeds of MVs and NMVs were

estimated for the test traps as mentioned

before. These, with respect to the total flow on

the road segment, are shown in Figure 4. The

impact of location and the frictions available at

that location was evident from the relative

scatter plots of the two categories of the

vehicles. Wide differences were observed in

the plots between MV and NMV speeds at

location 1, which is a two-lane and one-

directional section. The average speed of MVs

was found to be around 46 km/h, whereas, that

of NMVs was around 16 km/h. A gap of 30

km/h defined the road section category which

is NH. The presence of opposing direction

flow and the parking on road side at location 2

reduced the gap between the speeds of the two

categories. The average speed of MVs was

found to be 22 km/h and that of NMVs as 13

km/h, having a difference of only 9 km/h. The

impact of much higher frictions at road side at

location 3, as already mentioned, reduced the

gap to only 5 km/h. The MVs moved at a

speed of 13 km/h and NMVs at 8 km/h.

The average speeds, their standard

deviations and range observed at each of the

test location are given in Table 3.

Location 1 Location 2 Location 3

Figure 3: Space occupied by NMVs from side of carriageway


30

Table 3: Speed distribution profile for MVs and NMVs at three locations

Data collection

location

Average speed

(km/h)

Standard deviation

(km/h)

Range (km/h)

MVs NMVs MVs NMVs MVs NMVs

Location 1 45.9 16.315 5.71 1.4 36.45-57.9 19.61-13.02

Location 2 21.98 12.649 2.9 .7355 15.6-30.5 14.81-10.69

Location 3 12.85 8.48 2.17 .9149 8.0-18.89 10.28-6.142

Figure 5 shows the reduction in the speed

of MVs and NMVs due to increase in the

friction. It was noted that the reduction in MV

speeds was following negative exponential.

Higher reduction was observed with the

presence of opposing traffic and parking of

vehicles on road / carriageway side. It was 52

% between location 1 and 2 and 41.5%

between location 2 and 3, whereas, in the case

of NMVs it was found to be 22.5% and 33%

respectively.

Figure 5: Reduction in speeds due to presence of

roadside frictions

2.8 Speed-Flow Variations

The speed-flow data were plotted for the

three locations. As mentioned before, the flow

was taken in different forms, namely NMV

flow, total flow and opposing flow. Total flow

considered MV and NMV flow in the direction

of movement and opposing traffic flow was

considered at location 2 and 3 (both catering to

two-directional flow).

Figure 6 presents the plot between NMV

speed and flows for location 1, 2 and 3

respectively. Behavioural changes were

observed when the three plots having similar

parameters were compared across the

locations. Location 1, being catering to one-

direction flow, sufficiently wide and having

negligible friction on sides, showed an

increase in the NMV speeds with respect to the

NMV flow, but a reduction in the NMV speed

if total flow was considered. The increase in

the side frictions and use of same width by

both directional flows, as depicted by location

2, showed an increase in the NMV speeds at

lower rate with respect to the NMV flow, as

(a) Location 1 (b) Location 2 (c) Location 3

Figure 4: MV and NMV speed vs total flow


31

well as with the total flow. But the presence of

opposing flow and an increase in it caused a

reduction in the NMV speeds. Another point

of observation was the lower dispersion of

speed data at location 2 with respect to the

location 1. The plot of location 3, which

represented the higher level of frictions and

space and flow constraints, indicated that even

with lower flow values the speeds of NMVs

were low, and these reduced with an increase

in the NMV, total and opposing flows.

Same approach was used to analyze the

speeds of bicycles with a change in the flow

values (under different forms) and level of

frictions imposed by the road side

developments. Bicycle flow values were also

considered along with the NMV, total and

opposing direction flow as per location. These

are shown in Figure 7. The speeds were found

to be influenced more by road side frictions as

compared to the flow constraints or increase in

the flow. Bicycle speeds were found to be

increasing with bicycle and NMV flow at

location 1, but looked somewhat unaffected at

location 2 and found decreasing at location 3.

The impact of total flow and opposing flow on

bicycle speeds was relatively more at location

3 as compared to the other two locations. The

variations in cycle rickshaw speeds with flows

are shown in Figure 8.

(a) Location 1

(b) Location 2

12

13

14

15

16

17

18

19

1300 1500 1700 1900 2100 2300

NM

Vs

SPEE

D (

km/h

)

TOTLA FLOW (pcu/h)

12

13

14

15

16

17

18

19

300 500 700 900

NM

Vs

SPEE

D (

km/h

)

NMVs FLOW (pcu/h)

89

10111213141516

200 700 1200

NM

Vs

SPEE

D (

km/h

)

NMVs FLOW (pcu/h)

89

10111213141516

400 900 1400

NM

Vs

SPEE

D (

km/h

)

TOTAL FLOW (pcu/h)

89

10111213141516

400 900 1400

NM

Vs

SPEE

D (

km/h

)

CONFLICTING FLOW (pcu/h)


32

(c) Location 3

Figure 2: Variation in NMV speed with flows

(a) Location 1

(b) Location 2

6

7

8

9

10

11

12

200 400 600 800 1000

NM

Vs

SPEE

D (

km/h

)

NMVs FLOW (pcu/h)

6

7

8

9

10

11

12

600 800 1000 1200

NM

Vs

spee

d (

km/h

)

TOTAL FLOW (pcu/h)

6

7

8

9

10

11

12

600 800 1000 1200

NM

Vs

SPEE

D (

km/h

)

CONFLICTING FLOW (pcu/h)


33

(c) Location 3

Figure 3: Variation in bicycle speed with flows

(a) Location 1

(b) Location 2


34

(c) Location 3

Figure 4: Variation in cycle rickshaw speeds with flows

Cycle rickshaw speeds were found varying

in a range with respect to the flows, the

combined trend indicating lesser influence of

the locational factors as per the observations

made at location 1 (negligible frictions) and

location 2 (some frictions). Higher level of

frictions and restrictions of space on the

carriageway caused a reduction in the speeds

of the cycle rickshaws with an increase in the

flow values.

2.9 Microscopic Speed Analysis

Micro-level analysis of NMV speeds was

carried out with respect to the variations in the

flow values (in different ranges) and percent

share of NMVs in the total flow. These

variations are shown in Figure 9 for all the

three locations. Under one-directional flow on

a two-lane carriageway with negligible side

friction (Location 1), the NMV speeds were

found increasing with an increase in NMV

share in the total flow of upto 1800 pcu/h.

Above this value of flow the NMV speeds

were found decreasing even with an increase

in the percent share of NMVs. On such a road

section, if NMV share remains below 30%,

then even with an increase in flow the NMV

speeds increased. With an increase in side

friction and traffic moving in both the

directions (Location 2), the NMV speeds were

found increasing with an increase in percent

share of NMVs for all flow values. An

increase in the NMV share above 65% caused

reduction in NMV speeds.


35

(a) Location 1

(a) Location 2

(c) Location 3

Figure 5: Variation in NMV speeds with flow and percent share

121314151617181920

20 30 40 50

NM

Vs

SPEE

D (

km/h

)

PERCENTAGE NMVs SHARE IN TOTAL FLOW

BELOW 1500 pcu/hr 1500 pcu/hr TO 1800 pcu/hr

ABOVE 1800 pcu/hr

121314151617181920

1300 1500 1700 1900 2100 2300

NM

Vs

SPEE

D (

km/h

)

TOTAL FLOW (pcu/h)

BELOW 30% NMVS 30% NMVS TO 35% NMVS

ABOVE 35% NMVS

10

11

12

13

14

15

45 55 65 75

NM

Vs

SPEE

D (

km/h

)


below 800 pcu/hr 800 pcu/hr to 1000 pcu/hr above 1000 pcu/hr

10

11

12

13

14

15

400 600 800 1000 1200 1400

NM

VS

SPEE

D (

km/h

)

TOTAL FLOW (pcu/h)

below 55% NMVs 55% NMVs to 65% NMVs above 65% NMVs

6

7

8

9

10

11

50 60 70 80

NM

Vs

SPEE

D (

km/h

)


BELOW 750 pcu/hr 750 pcu/hr TO 950 pcu/hr ABOVE 951 pcu/hr

6

7

8

9

10

11

600 800 1000 1200

NM

Vs

SPEE

D (

km/h

)

TOTAL FLOW (pcu/h)

below 65% NMVs 65% NMVs to 70% NMVs above 70% NMVs


36

Further increase in the road space and flow constraints (Location 3), in general, showed a reduction in the NMV speeds even with an

increase in the NMV shares.

4.0 CONCLUSIONS

The study concludes that IRC: 11 – 1962,

“Recommended practice for the design and

layout of cycle tracks” does not provide any

worthwhile and relevant information on the

flow characteristic of NMVs. Very little

research has been conducted related to the

NMV traffic and its flow characteristics in the

urban areas. This study tried to provide some

input in the area of speed characteristics of

NMVs. Some of the findings are as follows:

i. High impact of road space and flow

constraints was observed on the speeds of

MVs and NMVs. This was negative

exponential for MVs and almost linear

reduction in the case of NMVs. Under the

worst conditions of frictions and

constraints, the two speeds became quite

close to each other. This indicated that

under the constraints NMVs are equally

efficient as MVs. They keep operating with

lower reduction in speeds as compared to

MVs.

ii. The speed-flow diagrams for the three

locations indicated that with the reduction

in car and heavy vehicle traffic on a road

segment, the NMV speeds picked-up even

with increasing total flow. But the increase

in the constraints affected them negatively.

The presence of opposing flow caused a

reduction in NMV speeds irrespective of

the level of friction available at a location.

The speed-flow relationship for the NMVs

cannot be ascertained. This might be due to

dispersion of values in close proximity to

each other. Looking at this condition an

envelope form is suggested to describe the

range of values within which the speed of

the NMVs can vary for different flow

values.

iii. The increase in the percent share of NMVs

caused an increase in the NMV speeds, on

both one-directional and two-directional

sections and without or with some friction,

for flows upto 1800 pcu/h. Above this flow

a reduction was observed in the NMV

speeds. Reducing tends were found on

sections with high frictions even if the

NMV share was quite high in the total

traffic. This indicated that space and flow

constraints dominate at even low traffic

volumes with high share of NMVs.

iv. The study has demonstrated that NMV

flows are affected more by frictions and

constraints rather than traffic volumes.

Their share in the traffic volume also

impacts the flow characteristics.

The study indicated towards provision of

dedicated space allocation to NMVs on roads.

Under normal traffic flow conditions and

negligible side frictions, 2.0 m space was

found adequate enough to cater to heavy flows

of NMVs. Higher side frictions caused the

NMVs to shift towards the centre of the

carriageway. Proper enforcement of road side

parking outside shoulders and management of

shops at the road side might help in restricting

them to the marked side strip of 2.0 m from

edge.

REFERENCES

1. American Association of State Highway

and Transportation Officials, “Guide for

the Development of Bicycle Facilities”

(1999), American Association of State

Highway and Transportation Official,

Washington, DC.

2. Duthie, J., Brady, J. F., Mills, A. F.,

Machemehi, R. B. (2010), “Effect of On-

Street Bicycle Facility Configuration on

Bicyclist and Motorist Behavior”, Journal

of Transportation Research Record, No.

2190, 37-44.

3. El-Geneidy, A., Krizek, K. J., Iacono, M.

(2013), “Predicting Bicycle Travel Speeds

along different Facilities using GPS Data:

A Proof of Concept Model”, Journal of

Transportation Research Part D: Transport

and Environment, Vol. 16, No. 2, 172-177.


37

4. Hongwei,G., Ziyou, G., Xiaomei, Z.,

Xiaobao, Y., (2011), “Traffic Behavior

Analysis of Non-motorized Vehicle under

Influence of Curb Parking”, Journal of

Transportation System Engineering and

Information Technology” , Vol. 11, No. 1,

79-84.

5. IRC :11 – 1962, “Recommended Practice

for the Design and Layout of Cycle

Tracks” , Indian Roads Congress, New

Delhi, India.

6. Navin, Francis P.D. (1994), “Bicycle

Traffic Flow Characteristics: Experimental

Results and Comparisons”, Journal of ITE.

Vol. 64, No. 3, Institute of Transportation

Engineers, Washington, DC, 31-36.

7. Wang, H., Wu, T., (2003), “A New United

Microcosmic Model of Urban Mixed

Traffic Flow”, Journal of IEEE, 156-162

URBAN TRANSPORT

JOURNAL


* Assistant Professor, School of Management Studies, Motilal Nehru National Institute of Technology (MNIT),

Allahabad, U. P. (India), [email protected]

** Research Scholar, School of Management Studies, MNIT, Allahabad, U. P. (India), [email protected]

SERVICE QUALITY DETERMINANTS FOR PUBLIC TRANSPORT

AND USE INTENTION: A STUDY OF COMMUTERS AND NON-

COMMUTERS IN INDIA

Dr. Vibhuti Tripathi* and Gunjan Nema**

Abstract: Service Quality is one of the important practical themes for transport service providers and regulatory

agencies. A better service quality shapes and improves commuters’ intentions to use public transport services

and also encourage the non-commuters to use the services in turn leading to enhanced productivity, profitability

and environment protection due to increased usage. Identification of service quality determinants would

influence the service improvement initiatives. The paper attempts to address the prime objective of identifying

the service quality of determinants for commuters and non-commuters and their influence on use intention. This

paper is based on empirical studies conducted in the cities of Delhi, Mumbai, Allahabad and Jabalpur amongst

the commuters and non-commuters. It was found that both the groups gave weightage to Availability and

Tangibles factors as important service quality determinants though in different order along with Empathy and

Responsiveness for Commuters and Safety and Integration for non-commuters. These determinants may be

used as the guidelines to improve service quality further by governments or the service providers.

Key Words: Urban Public Transport, Service Quality, JnNURM.

1.0 INTRODUCTION

Development of transport is considered to

be the sine-quanon for the prosperity and

smooth functioning of an economy. The

modern society in its present form is

inconceivable without the development of

rapid transportation systems. Rapid growth of

population in the cities, the appearance of

large manufacturing activities, the fast

urbanization of various territories have all

contributed to the development of various

modes of transport world over (Allen &

Thomas, 2000). The economic, social as well

as political progress of a country exclusively

depends on the progress of transport system,

making transport as an essential component of

basic infrastructure in modern era. A good

urban transport helps to promote urban

economy, enables social interactions, increases

productivity of resources, provides mobility to

people, enables accessibility to opportunities,

and sets directions and pattern of growth

(Ranganathan N, 1999). According to World

Bank Report on, ‘A study on urban transport

development’, the role of urban transport can

be described in a wider context by focusing on

issues like inputs for efficient urban

development, determinant of the quality of

urban life, and as an essential service to the

urban poor (The World Bank, 2000).

Poor transport systems stifle economic

growth and development and the net effect

may be a loss of competitiveness in both

domestic as well as international markets

(Padam, S., 2001). The impacts of a poor

urban transport system is manifested in terms

of congestion, delays, accidents, high energy

consumption, low productivity of resources,

high pollution to the environment, inequitable

access to services (Ranganathan N,1999) and

reduced service quality. The combination of

rapid urbanization and motorization has been a

key cause of numerous transport problems in

developing cities in Asia. It has resulted in a

deterioration in accessibility, service levels,

safety, comfort, operational efficiency, and the

urban environment (A study on urban transport

development, The World Bank USA, August

2000).

Service Quality Determinants for Public Transport and Use Intention: A Study of Commuters and

Non-Commuters in India

40

The usage of public transport in India has

been declining over the recent years despite

the fact that travel demands have increased

significantly. The causes for this decline in

India can be attributed to the facts like

underdeveloped infrastructure facilities, poor

quality of services and socio-cultural stigmas

associated with the use of public transport. The

Urban Transport System in India lacks the

quality and accessibility to match the

expectations of passengers and is far behind

the international standards. Service delivery of

Urban Transport in India is mainly supply-

oriented and capital intensive there is a need to

develop a market-based and customer-oriented

approach to change transport infrastructure

services. Satisfaction with services is not

rising in line with delivery improvements

undertaken through policy initiatives by

Government of India. This situation is referred

as ‘delivery paradox’ by Eboli L, Mazzulla G,

(2008). According to Low and Gleeson, (2003)

customer orientation and localized needs

should be the priority while planning public

services to be offered. If the service quality is

lacking and not well implemented, it will lead

to negative perception and dissatisfaction

(Karen and Peter, 2007). Research shows that

satisfaction with public services has remained

static; despite service improvements. Service

quality measure thus is a subject of great

interest both for planners and transit operators

(Eboli L, Mazzulla G, 2008).If the challenges

related to service delivery are addressed

effectively, it would lead to more efficient use

of service resources, increased profitability,

improved customer retention, increased

customer trust, reduced costs per customer,

and reduced turnover. This will also lead to

greater passenger satisfaction and less resource

wastages on unnecessary improvements and

may be a possible answer to the delivery

paradox (Blaug et al., 2006). Truly sustainable

transportation has not been achieved in any

region of the world. It is one of the important

practical themes for service providers and

regulatory agencies, but it also continues to be

a challenging research theme (Tripathi V. et al.

2012). While the service quality determinants

have been identified by researchers

internationally, a valid model in the Indian

context needs to be developed for facilitating

the service improvement initiatives. A better

service quality will shape and improve

commuters’ intentions to use public transport

services and also encourage the non-

commuters to use the services. The paper

attempts to address the objectives of:

1. Identifying Service Quality Determinants in

Commuters and Non-commuters.

2. Exploring the relationship of Service

Quality Determinants and Usage Intention.


Service quality as a concept has stimulated

an extensive interest and deliberation in varied

fields of services due to the complexities

posed in measuring and assessing it

(Wisniewski, 2001). It can act as a significant

differentiator for any service provider

(Parasuraman and Zeithaml, 1988). Proponents

of this concept initially defined it “as the

degree to which a customers’ perception of the

service encounter equates or exceeds their

expectations for the service” (Parasuraman et

al., 1985: 18). Alok (2013) defined service

quality as the extent to which the service, the

service process and the service organization

can satisfy the expectations of the user.

Service quality is of essence to any industry

because it has an effect on customer purchase

behaviour and retention (Oh & Mount, 1998).

Generally before a customer evaluates service

quality, he/she equate the service received

with what he/she expected (Voss,

Parasuraman& Grewal, 1998). According to

Fitzsimmons and Fitzsimmons (1998), there is

quality service delivery when perceptions

exceed expectations, satisfaction when

expectations are met and unacceptable

(negative) service quality when expectations

are not met.



41

The provision of good public transport

services is critical in alleviating the negative

impacts and achieving sustainability’s triple-

bottom-line goals, namely environmental,

economic and social goals (Too and Earl,

2009). The measures of public transport

service quality have been piecemeal. The

important questions relate to establishing a

framework for measuring public transport

service quality, identifying the top priorities

for improving user satisfaction levels with

public transport (Too and Earl, 2009),

measuring specific requirements of different

commuter and non-commuter groups, city

specific requirements and measuring specific

requirements during different times of the day

and locations etc make the entire gamut of

service quality delivery in public

transportation services a complex issue. This

poses a need to identify array of factors or

determinants that may influence the service

quality of a public transport in India. A

number of researchers have provided lists of

quality determinants, but the most

comprehensive and widely used model of

measuring service quality is the SERVQUAL

scale proposed by Parasuraman et al. (1985,

1988)based on the disconfirmation of

expectations model (Oliver, 1980) is widely

used to measure service quality. Parasuraman

et al. (1993) hold the view that their

SERVQUAL items are the basic skeleton

underlying service quality that can be

supplemented with context specific items

when necessary. A review of literature shows

that SERVQUAL (with the 5 RATER

dimensions of Reliability, Assurance,

Tangibles, Empathy and Responsiveness) has

emerged as the predominant model for

measuring service quality not only for services

in general, but also in the public transport

service industry.

Allen and DiCesare (1976) considered that

quality of service for public transport industry

contained two categories: user and non – user

categories. Under the user category, it consists

of speed, reliability, comfort, convenience,

safety, special services and innovations. For

the non –user category, it is composed of

system efficiency, pollution and demand.

Sillock (1981) conceptualized service quality

for pubic-transport industry as the measures of

accessibility, reliability, comfort, convenience

and safety. According to Middleton (1998a)

service quality in public transportation system

constitutes of internal and external factors

which affect the commuter’s perception

towards the public transport services. Internal

factors such as strategic issues (Lee, Lee, &

Lee,2006), top management commitment,

service quality standards (Middleton, 1998b),

monitoring systems (Deegan, 2002; Gray,

2002; Alexandre & Short, Dec 1995/Jan

1996), customer complaints handling system

(Kotler &Kavin, 2008) and external factors

such as alternative services (Evans & Shaw,

2001; Michel, 1999), frequency of traveling

and timings (Flem&Schiermeyer, 1997;

Galetzka, Gelders, Verckens, &Seydel, 2008)

convenience and comfort (Regis, 1996),

climate, ego, social status,

professions(Sanchez, 1999)

Various dimensions studied by researcher

to measure service quality in public transport

are summarized in Table 1.

Table 1: Different Dimensions Used to Measure Service Quality in Public Transport

S.No Author and year Study Dimensions

1 Silcock,1981 Public transport industry Accessibility, reliability, comfort,

convenience and safety

2 Hanna and

Drea,1998

Rail passenger service

quality (Amtrack in US)

comfort, timing, cost, location, in transit

productivity

3 Drea and

Hanna,2000


quality (Amtrack in US)

Non servqual (cost, convenience getting

to station, parking availability, comfort,



42

S.No Author and year Study Dimensions

seat comfort, ride, seating area

cleanliness, service of on board staff)

4 Tripp and

Drea,2002


quality

Non servqual (announcements, seat

comfort, ride, cleanliness of seating

area, courtesy of on board staff, rest

rooms, café car conditions)

5 Cavana RY,

Corbett LM and

Lo YL ,2005(3

column servqual)

Rail service quality,

(Wellington, New Zealand)

Servqual with modification (RATER+

comfort, connection and convenience)

6 LelL,Mac L,2005 Transport service

sector(South China)

RATER+ loyalty

7 Eboli L, Mazzulla

G,2008

Public transport(Italy) Stated preference experiment for

measuring service quality

8 Too L Earl

G,2009

Public transport(Australia) SERVQUAL with the following factors

-tangibles, responsiveness, reliability

and assurance.

9 Prasad MD,

Shekhar BR,2010

a

Service quality, Indian

Railways

Servqual with modification(Zone of

tolerance study) RATER+ comfort

10 Prasad MD,

Shekhar BR,2010

b

Service quality, Indian

Railways

RATER+ Service Product, Social

Responsibility and service delivery

11 Rita S, Ganesan

V,2010

Public transport (India-

Chennai)

A combination of SERVQUAL and

Kano model. 6 factors were Basic

services, Appreciative services,

Reliability, Assured services, Additional

services and Technological

advancements.

12 Randheer K,

Motawa A, Vijay

J,2011

Public transport India(first

study in India)

RATER+ culture (excluded Tangibles)

13 Sezhian M,

Muralidharan C,

Nambirajan T,

Deshmukh

SG(2011)

Public sector bus transport

company, India(SRTU-

Tamil Nadu)

customer expectations and company

responsibilities

Demographic characteristics and public

transport specific determinants shape the

service quality expectations they also

influence the generic dimensions. These

factors necessitate to develop type of

commuters or city specific measurement

scales. Karen and Boo (2007) have suggested

that the traditional SERVQUAL dimensions

may not be meaningful in all situations and

contexts. Svensson (2004) in his study has laid

the importance of customizing a particular

model to match the study context. Considering

the fact a measurement index for service

quality of public transport was developed by



43

the researchers using the basic five

SERVQUAL dimensions of Reliability;

Assurance; Tangibles; Empathy;

Responsiveness. As suggested by Parasuraman

et al. (1993). On analyzing the dimensions

used by researchers for measuring service

quality in public transport in the backdrop of

Indian public transport scenario and

demography of the country; it is observed that

Availability of public transport modes,

Integration of multiple modes, Affordability

and Safety also play a role in influencing the

service quality(Tripathi,V. et.al 2012),thus the

SERVQUAL was supplemented with 5 more

dimensions (Availability, Affordability,

Safety, Convenience, Integration) derived

through a pilot survey and a structured

interview schedule of experts.

3.0 RESEARCH METHODOLOGY

The theoretical framework implies

testable predictions about the service

quality determinants of commuters and non

commuters and their influence on use

intention. Two Specific predictions related to

the relationship are hypothesized as:

H1: Determinants of Service Quality in

Commuters have a direct relationship with use

intention.


Non-Commuters have a direct relationship

with use intention.

The modified instrument with 10

dimensions was tested for its applicability with

the citizens of two Tier 1 cities viz. Mumbai

and Delhi and two Tier 2 cities viz. Allahabad

and Jabalpur.

The sampling was in two stages- firstly

purposive sampling method was used to select

the cities on the basis of population,

implementation of JnNurm, modes of public

transport etc. The cities were chosen on the

criterion listed in the Table 2. In the second

stage the respondents were chosen through

convenience sampling method from the public

places like markets, bus stops, railway/metro

stations etc. A total of 685 responses were

collected.

Table 2: City Characteristics

TIER I Cities

Criteria Mumbai New Delhi

Population 19.6 million (Includes Greater

Mumbai UA and Vasai-Virar

Municipal Corporation)(The

Economic Times, New Delhi, 20 Oct

2011).

21.7 million (includes Delhi UA,

Faridabad, Gurgaon, Noida,Greater

Noida and Ghaziabad)(The Economic

Times, New Delhi,20 Oct 2011).

JnNURM

implementation

status

implemented implemented

Modes of public

transport available

Suburban trains and buses

(multimodal transport)

Buses, metro, ring railway

(multimodal transport)

Current public

transport usage

45% 43%

Proposed increase

in usage after

JnNURM

to 48% to 45%



44

TIER II Cities

Criteria Allahabad Jabalpur

Population 1,117,094(according to provisional

census data 2011)

1,054,336(according to provisional

census data 2011)

JnNURM

implementation

status

implemented implemented

Modes of public

transport available

Buses are the main modes of PT Buses are the main modes of PT

Current public

transport usage

10% 4%

Proposed increase

in usage after

JnNURM

to 12% to 10%

A survey was conducted using a structured

questionnaire both with the commuting and

non-commuting population. Respondents were

asked to rate their responses on a five point

Likert Scale ranging from Highly Important

(1) to Least Important (5). In order to address

the research objectives of the paper, the

collected data was analyzed with factor

analysis and regression analysis using SPSS

16. Collected data was subjected to skewness

and kurtosis study to determine normality. It

was inferred that the data is normally

distributed and thereafter Principal Component

Analysis was conducted separately for the data

collected for commuters and non-commuters

to determine the factors which shape the

perceptions of service quality for the two

groups separately. The derived factors were

considered as independent variables to further

test their relationship with dependent variable

(Use Intenstions) with the help of regression

analysis. Regression helps one understand how

the typical value of the dependent variable

changes when any one of the independent

variables is varied, while the other

independent variables are held fixed.

4.0 ANALYSIS AND FINDINGS a. Common Mode of Travel: The respondents

were asked to mention common mode

through which they commuted within the

city. On city wise analysis it was found that

despite of an established multi-modal

transport system in Delhi the use of public

transport as well private vehicles is equally

divided and is similar to the non- metro

cities of Jabalpur and Allahabad where

public transport system is yet to get fully

established. (Table 3).

Table 3: Most Commonly used Mode of Travel

City Public

Transport as

common mode

of travel (%)

Own Vehicle

as common

mode of

travel (%)

Mumbai 83.9 16.1

Delhi

NCR

50.0 50.0

Allahabad 49.7 50.3

Jabalpur 41.0 59.0

Total 56.6 43.4

b. Distance travelled per trip: On asking the

respondents to mention the distance they

travelled per trip it was found that

maximum percentage of (22.9%) travelled

more than between 5-10 kms. per trip. It is

also evident from Table 3 that trip distance

of the respondents was more in metro cities

in comparison to non-metro cities of

Jabalpur and Allahabad.



45

Table 4: City wise kilometers travelled per trip

City Less than 5

km (%)

5-10 km

(%)

10-15 km

(%)

15-20 km

(%)

20-25 km

(%)

More than

25 km (%)

Mumbai 2.0 4.7 5.1 4.5 2.6 7.3

Delhi NCR 1.5 3.1 5.3 6.1 4.4 4.5

Allahabad 2.0 8.5 5.5 3.8 2.0 2.8

Jabalpur 2. 6.7 5.3 3.4 3.1 3.6

Total 7.7 22.9 21.2 17.8 12.1 18.2

c. Determinants of Service Quality in

Commuters: Out of the total sample, the

data of 388 respondents was further treated

with Principle Component Analysis.

Bartlett’s test of sphericity and Kaiser-

Meyer Olkin (KMO) measures of sampling

adequacy were used to examine the

appropriateness of Factor Analysis. The

KMO statistic (0.942) is large and

significant (>.05), considering the value

Factor Analysis is considered as an

appropriate technique for further analysis

of data. KMO value closer to zero indicates

that a diffusion exists in the pattern of

correlations and suggests that factor

analysis will not be appropriate for the

sample (Field, 2009).As recommended by

Kaiser, 0.5 is the lowest threshold for

proceeding further. Values between .5 and

.7 as mediocre, upto .8 as good, upto .9 as

great and above .9 as excellent (Field,

2009). A significant Bartlett’s test

(p<0.001) indicates that correlations

between items are sufficiently large to

proceed with the PCA. KMO value is .942

and is excellent (Field, 2009).

Nine factors were extracted using the scree

plot criteria which explained 60.278%

variance in the data. According to Howitt

D, et.al it is more helpful to use the screen

test in order to estimate the number of

factors. Cronbach Alpha values were in the

range of .877 and .736 indicating high

internal consistency reliability.

Table 5: KMO and Bartletts Test (Commuters)

Kaiser-Meyer-Olkin Measure of

Sampling Adequacy. .942

Bartlett's Test

of Sphericity

Approx. Chi-

Square 10942.779

Df 1326.000

Sig. .000

Factor loadings in the range of .30 to .40

are considered to meet the minimal level for

interpretation of structure, .50 or greater are

considered practically significant. Loadings

above .70 are indicative of a well defined

structure (Hair et al, 2009).Individual items

were checked for factor loadings and the items

with loadings below .50 were dropped from

further analysis. Table 5 shows the factor

structure of commuters and the nomenclature

given to each of the extracted factor.

Table 6: Factor structure (Commuters)

S.no Factor Items Factor

Loadings

Cronbach

alpha

1 Empathy Employees are neat and well dressed. .662 .855

Driver, conductor and other employees are

courteous and helpful.

.632

Employees are prompt in responding to .601



46


Loadings

Cronbach

alpha

commuters.

The staff has enough knowledge and skill

to handle customer queries.

.581

2 Responsiveness There is provision to register complaints. .670 .877

Any complaint lodged is properly

addressed.

.764

Ample staff is available to handle requests

and complaints.

.755

First aid facilities are adequately available. .734

The frequency of breakdown is low and

there is enough back up service.

.518

Special services are planned according to

specific needs of various commuter groups

(old, disabled, women and children).

.531

3 Convenience Refreshment shops are in adequate

numbers.

.505 .830

Cleanliness is maintained at the

stops/stations.

.567

Proper lighting is maintained at the

stop/station.

.646

There are sufficient number of ticket

windows.

.583

Modernised ticket dispersal mechanism is

available.

.674

It is easy to use the ticket dispersal

mechanism.

.570

Proper ticket is issued in time. .557

4 Availability In case of changing routes there are

enough connecting public transport

options available.

.550 .863

The various public transport modes

available in the city are well integrated.

.571

In case I have to switch mode, I don’t have

to wait for long.

.553

Government public transport is available

in all parts of the city.

.737

Government public transport running on

different routes is sufficient.

.746


at all times of the day.

.695

5 Tangibles Parking facility near the boarding point is

adequately available.

.698 .736

In case I travel with my private vehicle to

the boarding point, I can easily park it

.676



47


Loadings

Cronbach

alpha

nearby.

The government public transport vehicles

are spacious.

.513

The public transport vehicles are well

maintained.

.549

There is enough provision to carry

luggage.

.537

Adequate seats are available to commuters

waiting at the boarding point of

government public transport.

.506

6 Affordability Tickets are affordable. .696 .753

Daily/monthly travel passes are available. .708

The passes are affordable. .669

7 Assurance Time schedules and fare charts are

displayed at adequate places.

.672 .829

Time schedules and fare charts are simple

to understand.

.676

Adequate number of announcements are

made.

.735

Any change in the time table is properly

communicated.

.653

8 Safety I feel safe while using the public transport

facilities.

.739 .850

I feel safe while using the public transport

facilities at night also.

.751

Night services are reliable. .682

In order to test the following hypothesis:


Commuters have a direct relationship with use

intention.

A stepwise multiple regression was

performed. The most important predictors

were entered stepwise and the other predictors

being non-significant are removed from the

analysis.

Table 7: Model Summary

Mode

l R

R

Squar

e

Adjusted R

Square

Std.

Error of

the

Estimate

Change Statistics

Durbin-

Watson

R Square

Change F Change

df

1 df2

Sig. F

Change

1 .454a .206 .204 .71033 .206 99.698 1 384 .000

2 .502b .252 .249 .69021 .046 23.723 1 383 .000

3 .538c .289 .283 .67397 .037 19.674 1 382 .000

4 .556d .310 .302 .66500 .021 11.373 1 381 .001 1.876

a. Predictors: (Constant), F1

b. Predictors: (Constant), F1, F5



48

c. Predictors: (Constant), F1, F5, F6

d. Predictors: (Constant), F1, F5, 6,

F4

e. Dependent Variable: DV

Table 8: Coefficients

Model

Unstandardized Coefficients Standardized Coefficients

B Std. Error Beta

Step 1 (Constant) 1.228 .129

Empathy .411 .041 .454

Step 2 (Constant) .964 .136

Empathy .306 .045 .338

Tangibles .189 .039 .244


Empathy .249 .046 .275

Tangibles .189 .038 .245

Responsiveness .166 .038 .201


Empathy .151 .054 .167

Tangibles .178 .038 .230

Responsiveness .137 .038 .166

Availability .175 .052 .195

R2 = .206, for Step 1, Δ R2 = .046 for Step 2, Δ R2 =.037 for Step 3, Δ R2 = .021 for Step 4

The standardized beta coefficients show the

relative impact on the dependent variable of a

change in 1 standard deviation in either

variable (Hair et al, 2007). The beta values

indicate the individual contribution of each

predictor to the model. The standardized

coefficient beta for model 4 for f1 is .167, for

f5 is .230, for f6 is .166 and for f4 is .195.The

corresponding t values for all these predictors

are significant at (p<.05) as per the

recommendations of (Field, 2009).All

predictors are hence significant predictors of

the model.

The following hypothesized relationships were

assessed using Stepwise Multiple Regression:



49

Hypothesis Predictor/Independent

variable

Direction of

relationship

Dependent

variable

Regression analysis

for hypothesis

H1a Empathy → Use Intention Supported

H1b Responsiveness → Use Intention Supported

H1c Convenience → Use Intention Not Supported

H1d Availability → Use Intention Supported

H1e Tangibles → Use Intention Supported

H1f Affordability → Use Intention Not Supported

H1g Assurance → Use Intention Not Supported

H1h Safety → Use Intention Not Supported

The following regression model is obtained for commuters-

d. Determinants of Service Quality in Non-

Commuters: Out of thetotal sample, the

data of 279 respondents was further

treated with Principle Component Analysis.

Bartlett’s test of sphericity and Kaiser-

Meyer Olkin (KMO) measures of sampling

adequacy were used to examine the

appropriateness of factor analysis. The

KMO statistic (0.901) is large and

significant (>.05), a significant Bartlett’s

test (p<0.001) indicates that correlations

between items are sufficiently large to

proceed with the PCA.

9 factors were extracted using the scree plot

criteria which explained 60.278% variance

in the data. (Howitt D, Cramer D, 2011)

say that it is more helpful to use the scree

test in order to estimate the number of

factors. Cronbach Alpha values were in the

range of .695 and .866 indicating high

internal consistency reliability.

Table 9: KMO and Bartletts Test (Non-

Commuters)

Kaiser-Meyer-Olkin

Measure of Sampling

Adequacy.

.901

Bartlett's Test of

Sphericity

Approx.

Chi-

Square

7553.064

Df 1326.000

Sig. .000

Table 9 shows the factor structure of

Commuters and the nomenclature given to

each of the extracted factor.



50

Table 10: Factor structure (Non-Commuters)


loadings

Cronbach

alpha

1 Tangibles The government public transport vehicles

are spacious. .504 .866

The public transport vehicles are well

maintained

.539

Seats in the government public transport

vehicles are comfortable. .561

There is enough provision to carry

luggage. .659

The government public transport vehicles

are punctual. .661

Cleanliness is maintained at the

stops/stations. .654

Proper lighting is maintained at the

stop/station. .704

In case I travel with my private vehicle to

the boarding point, I can easily park it

nearby.

.706

Parking facility near the boarding point is

adequately available. .711

The boarding points are well maintained. .614

There are sufficient numbers of ticket

windows. .555

2 Integration

In case of changing routes there are

enough connecting public transport

options available.

.674 .770

The various public transport modes

available in the city are well integrated. .669

In case I have to switch mode, I don’t

have to wait for long. .644

The government public transport stops for

sufficient time for boarding and

unboarding.

.551

It is easy to board and unboard the

bus/train. .635

3 Availability Government public transport is available

in all parts of the city. .737 .863

Government public transport running on

different routes is sufficient. .746


at all times of the day. .695



51

In order to test the hypothesis


Non-Commuters have a direct relationship

with use intention.

A stepwise multiple regression was

performed to ascertain the important

determinants of public transport service

quality among non-commuters. The following

relationship was thus revealed.

4 Empathy Employees are neat and well dressed. .519 .715

Driver, conductor and other employees

are courteous and helpful. .563

5 Responsive-

ness There is provision to register complaints. .510 .812

Ample staff is available to handle requests

and complaints. .619

First aid facilities are adequately

available. .710

The frequency of breakdown is low and

there is enough back up service. .661

Special services are planned according to

specific needs of various commuter

groups (old, disabled, women and

children).

.517

6 Affordability Tickets are affordable. .664 .695

Daily/monthly travel passes are available. .684

The passes are affordable. .716

8 Assurance Time schedules and fare charts are

displayed at adequate places. .643 .773

Time schedules and fare charts are simple

to understand. .636

Adequate number of announcements are

made. .763

Any change in the time table is properly

communicated. .722

9 Safety I feel safe while using the public transport

facilities. .712 .787

I feel safe while using the public transport

facilities at night also. .718

Night services are reliable. .559



52

Table 11: Model Summarye

Model R R2 Adjusted

R2

Std. Error

of the

Estimate

Change Statistics Durbin

-

Watson

R2

Change

F

Change

df1 df2 Sig. F

Change

1 .532a .283 .281 .83631 .283 116.701 1 295 .000

2 .599b .358 .354 .79280 .075 34.271 1 294 .000

3 .621c .385 .379 .77718 .027 12.934 1 293 .000

4 .638d .406 .398 .76507 .021 10.351 1 292 .001 1.759

a. Predictors: (Constant), f7

b. Predictors: (Constant), f7, f2

c. Predictors: (Constant), f7, f2, f8

d. Predictors: (Constant), f7, f2, f8,

f3

e. Dependent Variable: dv

Table 12: Coefficientsa

Model Unstandardized Coefficients Standardized Coefficients

B Std. Error Beta

1 (Constant) 1.494 .155

Safety .495 .046 .532

2 (Constant) .873 .181

Safety .321 .053 .345

Assurance .353 .060 .332

3 (Constant) .570 .196

Safety .305 .052 .328

Assurance .312 .060 .293

Tangibles .166 .046 .172

4 (Constant) .411 .199

Safety .277 .052 .298

Assurance .231 .064 .217

Tangibles .160 .045 .166

Integration .171 .053 .175

The following hypothesized relationships were assessed using Stepwise Multiple Regression

Hypothesis Predictor /

Independent variable

Direction of

relationship

Dependent

variable

Regression analysis

for hypothesis

H2a Tangibles → Use Intention Supported

H2b Integration → Use Intention Supported

H2c Empathy → Use Intention Not Supported

H2d Responsiveness → Use Intention Supported

H2e Affordability → Use Intention Not Supported

H2f Assurance → Use Intention Not Supported

H2g Safety → Use Intention Supported

The following regression model was obtained for non-commuters-



53

The predictors in the model explain 39.8%

variance in the dependent variable

iebehavioral intentions.

5.0 CONCLUSION

The study reveals that barring Mumbai

which has a well established and integrated

public transport in all other cities of Delhi,

Allahabad and Jabalpur respondents used

public transport and private vehicles equally. It

can be attributed to the fact that the

accessibility of Public Transport in the city

and sub-urban areas is limited to certain parts

only also that the public transport system is yet

to get further developed by improvements in

infrastructure and networks in comparison to

Mumbai, where the share of respondents using

public transport is 83.9%.

The study indicates that the commuters

assign more weight to Empathy (Employee

Related Aspects etc.) followed by Tangible

(Condition of the Vehicles, Boarding Points

etc.), Responsiveness (Complaints Handling,

Special Services etc.) and Availability

(Sufficient routes, Available all times etc etc.)

factors. Whereas other factors of Convenience,

Assurance, Affordability and Safety are not as

important service quality determinants. Non-

commuters pay maximum importance to

Integration (Different Routes, Connecting to

different modes etc.) while the next important

was Availability (Sufficient routes, Available

all times etc) followed by Tangibles

(Condition of the Vehicles, Boarding Points

etc.) and Safety factors (Safety while using

public transport, safety during Night use etc.).

Both the groups of commuters and non-

commuters indicate a separate set of service

quality determinants; they give importance to

Availability and Tangibles though in a

different order. Empathy and Responsiveness

are other two important determinants of

service quality for commuters. Safety and

Integration are the other important service

quality determinants for Non-commuters.

It can be concluded that public transport

service providers need to understand and

provide reliable services to the commuters

consistently. A better service quality will

shape and improve passenger intentions to use

public transport services and attract a large

number of people to minimize the use of

privately owned transport. Identification of

service quality determinants influences the

service improvement initiatives. When

effective and efficient systems are put in place

gradually to monitor the service quality

determinants the desired goal of providing

quality of service can be achieved which will

also address issues of urban pollution and

traffic congestion in most of the cities. These

determinants may be used as the guidelines to

improve service quality further by

governments or the service providers.

REFERENCES 1. Alexandre, A., & Short, J. (Dec 1995/Jan

1996). Can pricing change urban travel?

Organization for Economic Cooperation

and Development. The OECD Observer,

197, 20-21.

2. Allen, W.G., &Dicesare, F. (1976). Transit

Service evaluation: preliminary



54

identification of variables characterizing

level of service. Transportation Research

Record, Vol. 606, pp 47-53

3. Allen & Thomas (2000). Poverty and devel

opment into the 21st Century, Oxford:

Oxford. University Press.

4. Alok, K. R. (2013). Customer relationship

management: Concepts and cases (2nd

ed.). New Delhi, India: PHILearning.

5. Andy Field (2007). Discovering Statistics

using SPSS, SAGE Publications Ltd, ISBN

1446242242, 9781446242247

6. Blaug, R., L. Horner & R. Lekhi (2006a),

Public Value, Citizen Expectations and

User Commitment. A Literature Review.

London: Work Foundation.

7. Cavana RY, Corbett LM & Lo YL (2007).

Developing zones of tolerance for

managing passenger rail service quality”,

International Journal of Quality and

Reliability Management, Vol 24, No

1,2007,pp 7-31

8. Deegan, C. (2002). The legitimizing effect

of social and environmental disclosures: a

theoretical foundation. Accounting,

Auditing and Accountability, 15(3), 282-

311.

9. Drea, J.T. & Hanna, J.B. (2000). Niche

marketing in intrastate passenger rail

transportation, Transportation Journal,

Vol. 39 No. 3, pp 3343

10. Eboli L, &Mazzulla G,(2008). Willingness-

to-pay of public transport users for

improvement in service quality, European

Transport \ TrasportiEuropei n. 38 (2008):

107-118.

11. Evans, G., & Shaw, S. (2001). Urban

leisure and transport: Regeneration effects.

Journal of Leisure Property, 1(4), 350-372.

12. Fitzsimmons, J. A., & Fitzsimmons, M.

J.(1998). Service Management. New York:

McGraw-Hill.

13. Flem, H. L., &Schiermeyer, C. (1997). Rail

transit: The people's choice. Railway Age,

198(9), 71-80.

14. Galetzka, M., Gelders, D., Verckens, J. P.,

&Seydel, E. (2008). Transparency and

performance communication: a case study

of Dutch Railways. Corporate

Communications: An International Journal,

13(4), 433-447.

15. Gray, R. (2002). The social accounting

project and accounting organizations and

society Privileging engagement,

imaginings, new accountings and

pragmatism over Critique? Accounting,

Organizations and Society, 27, 687-708.

16. Hair,Black,Babin,Anderson,Tatham(2009).

Multivariate Data Analysis, Sixth

Edition,Pearson Education.

17. Hanna, J.B. &Drea, J.T. (1998).

Understanding and predicting passenger

rail travel: an empirical study,

Transportation Journal, Vol. 38 No. 1, pp

3846.

18. Howitt D, Cramer D (2011). Introduction

to SPSS Statistics in Psychology-For

version 19 and earlier, Fifth Edition,

Pearson Education Limited.

19. Karen T, Peter S (2007). An investigation

of the relationship between public transport

performance and destination satisfaction,

Journal of Transport Geography, 15 (2007)

136–144,

www.elsevier.com/locate/jtrangeo

20. Karen, K., and Boo, H. V. (2007). Culture

and service quality expectations Evidence

from Generation Y consumers in Malaysia.

Managing Service Quality, 17(6), 656-680.

21. Kotler, P., and Kavin, L.K. (2008).

Marketing Management. (13th ed). United

States: Pearson Education Inc.

22. Lee, S., Lee, S., and Lee, Y.I. (2006).

Innovative Public Transport Oriented

Policies in Seoul. Transportation

Amesterdam, 33(2), 189-204.

23. Lei Mandy & Mac Lancy (2005). Service Quality

And Customer Loyalty In A Chinese Context:

Does Frequency Of Usage Matter?

ANZMAC 2005 Conference: Services

Marketing, pp 138-145.

24. Low N. & Gleeson B(2003), Making Urban

Transport Sustainable. Palgrave

Macmillan: New York.

http://www.elsevier.com/locate/jtrangeo



55

25. Michel, T. (1999). The taxi as a transit

mode. Transportation Quarterly.

Lansdowne, 53(4), 121-130.

26. Middleton, W.D. (1998a). Why some voters

say yes; why some say no. Railway Age,

199(2), G1-G2.

27. Middleton, W.D. (1998b). Urban rail

continues to grow. Railway Age, 199(2),

G10-G19.

28. Nutsugbodo, R Y(2013). Tourists’

perceptions of the quality of public

transportation services in the Accra

metropolis: a Servqual approach, African

Journal of Hospitality, Tourism and

Leisure Vol. 2 (4) - (2013)

29. Oh, H., & Mount, D. J. (1998). Prediction

of brand loyalty with service unit

performances in lodging operations.

Journal of International Hospitality,

Leisure, & Tourism Management, 1(3), 37-

54.

30. Oliver, R., (1980). A cognitive model of the

antecedents and consequences of

satisfaction decisions. Journal of

Marketing Research 17, 460–469.

31. Padam, S. (2001). Transport and Urban

Governance, Urban Transport Journal,

Vol. 2, No. 1, pp. 30-33.

32. Parasuraman, A., Berry, L., &Zeithaml, V.,

(1993). More on improving service quality

measurement, Journal of Retailing, 69,

140-7.

33. Parasuraman, A., Zeithaml, V. & Berry,

L.L. (1988). SERVQUAL: a multiple-item

scale for measuring consumer perceptions

of service quality, Journal of Retailing, Vol.

64, Spring, pp. 12-40.

34. Parasuraman, A., Zeithaml, V., & Berry,

L.L. (1985). A conceptual model of service

quality and its implications for future

research. Journal of Marketing, 49(4), 41-

50.

35. Prasad MD, &Shekhar BR,(2010 b).

Impact of Service Quality Management

(SQM) Practices on Indian Railways - A

Study of South Central Railways,

International Journal of Business and

Management Vol. 5, No. 9; September 2010

36. Prasad MD,&Shekhar BR (2010 a).

Importance and the Zone of Tolerance of

Passengers’ Expectations of Indian

Railway Passengers Services by applying

RAILQUAL, Asian Journal of management

research, ISSN 2229 – 3795

37. Randheer, K., Al-Motawa, A. A., & Vijay,

P. J. (2011). Measuring commuters’

perception on service quality using

SERVQUAL in public transportation.

International Journal of Marketing Studies,

3(1), 21-34.

38. Regis, C. (1996). By bus, subway or cab.

Business Insurance. Chicago, 30(16), T18.

39. Ranganathan, N. (1995). National Urban

Transport Policy – A Framework, Indian

Journal of Transport Management, Vol. 19,

No. 2, pp. 85-98.

40. Rita S, &Ganesan V,(2010). Assessment of

QOS for Public Bus Transport System A

SERVQUAL Based Kano Approach,

Advances In Management. Vol. 3 (4) April

(2010)

41. Sanchez, W.T. (1999). The connection

between public transit and employment:

The cases of Portland and Atlanta.

American Planning Association. Journal of

the American Planning Association, 65(3),

284-296.

42. SezhianVetrivel, M., Muralidharan, C.,

Nambirajan, T. &Deshmukh, S. G. (2012).

Attribute-based perceptual mapping using

discriminant analysis in a public sector

passenger bus transport company: A case

study. J. Adv. Transp. doi: 10.1002/atr.200

43. Silcock, D.T. (1981). Measures of

operational performance for urban bus

services. Traffic Engineering and Control,

Vol. 22 No. 12, pp. 645-8.

44. Svensson, G, (2004). Interactive service

quality in service encounters: empirical

illustration and models. Managing Service

Quality, 14(4), 278–287.

45. Too Linda,& Earl George, (2009). Public

Transport Service Quality and Sustainable



56

Development: a Community Stakeholder

Perspective, Sustainable Development

Sust. Dev. 18, 51–61 (2010) Published

online 21 May 2009 in Wiley InterScience

(www.interscience.wiley.com) DOI:

10.1002/sd.412,Australia. Copyright ©

2009 John Wiley & Sons, Ltd and ERP

Environment

46. Tripathi V, Kumar A, &Nema G. (2012),

Measuring Service Quality in Indian Public

Transport System: A Comprehensive Review,

Asia-Pacific Marketing Review, Volume: I,

No.1, Jan-June 2012, pp. 113-119

47. Tripp, C &Drea, JT. (2002). Selecting and

promoting services encounter elements in

passenger rail transportation. Journal of

Services Marketing 16: 432–442

48. Voss, B. G., Parasuraman, A., & Grewal,

D. (1998). The role of price, performance

and expectations in determining

satisfaction in service exchanges. Journal

of Marketing,62(1), 46-61.

49. Wisniewski, Mik (2001). Using

SERVQUAL to assess customer satisfaction

with public sector services, Managing

Service Quality, vol. 11(6), p. 380-388

50. www.kas.de/upload/dokumente/megacities/

Urban_transport_development_asia-

2000.pdf A study on urban transport

development, The world bank USA, August

2000, accessed on 1/05/2014

URBAN TRANSPORT

JOURNAL


* Principal Scientist, Environmental Sciences Division, CSIR-Central Road Research Institute, New Delhi.

110025; [email protected]

** Scientist, Transportation Planning Division, CSIR-Central Road Research Institute, New Delhi. 110025;

[email protected]

*** Head, Transportation Planning Division, CSIR-Central Road Research Institute, New Delhi.

110025;[email protected]

GENERIC FRAMEWORK FOR ESTIMATING CARBON FOOTPRINT OF COMMUTING

WITH PUBLIC TRANSPORT MODES

Kirti Bhandari*, Mukti Advani**, Purnima Parida***

Abstract: Changes in climate caused by changes in anthropogenic (i.e. “man-made”) greenhouse gas (GHG)

emissions have become a major public policy issue in countries all over the world. With an estimated 28.4% of

these emissions attributed to the transportation sector, attention is being focused on strategies aimed at reducing

transportation GHG emissions. Quantifying the change in GHG emissions due to such strategies is one of the

most challenging aspects of integrating GHG emissions and climate change into transportation planning and

policy analysis. This research aims to develop a method for estimating the carbon footprint of commuting and

apply this method to the public transport systems existing in Delhi. A complete study on trip profile of the

transit commuters of available modes will be used to estimate the carbon foot prints for different mode-

combinational trips (trip profile including access, egress and main line haul mode). This methodology consists

of estimating the number of trips by each mode followed by estimating the direct CO2 emissions. Carbon

footprints provide insights into the potential impact of different policies. Questions such as where to apply

certain policies (both in terms of mode and geographic area) to gain the largest reductions can be answered

using such footprints.

Keywords: Framework, access, egress, trips, emission factors, CO2 emissions

1.0 BACKGROUND

Transport sector contributes around 14%

towards the global emissions of greenhouse

gases [World Bank, 2011]. Carbon dioxide

represents the largest proportion of basket of

greenhouse gas emissions. During, the past

three decades, carbon dioxide emissions from

transport has increased faster than those from

all other sectors and are projected to increase

more rapidly in coming years, if no

intervention is done.. The road transport alone

emits around 16% of the global CO2 emissions

[IEA, 2007]. From 1990 to 2004, carbon

dioxide emissions from the world’s transport

emissions have increased by 36.5%.

As one of the most rapidly growing

countries and the fifth largest CO2 emitter in

the world, India is experiencing a rapid growth

in its economy as well motorized mobility

[OICA]. Passenger mobility in Delhi is poised

to increase at the rate of 8.7% reaching 534

billion pkm by 2020 [Bhandari, K. and Y.

Hayashi (2011)]. Buses form the backbone of

public transport, but remain very unreliable,

overcrowded and inefficient; this has resulted

in increased usage of personalized modes of

transport and its environmental consequences.

In this light the Government of India has

announced a national urban transport policy in

April 2006 as an integral part of the Jawaharlal

Nehru National Urban Renewal Mission

(JNNURM). The draft National Urban

Transport Policy (NUTP) aims at curtailing the

use of private vehicles and give impetus to

public transport and non-motorized vehicles.

The policy envisages encouraging 4 million

plus cities to plan for a mass transit system

adopting a technology that best suits the city

requirement. The options for this include buses

on dedicated corridors, elevated sky bus and

monorail systems, electric trolley buses and

metro systems.

2.0 INTRODUCTION

Worldwide, energy use is increasing faster

in the transport sector than in any other sector,

and fastest of all in developing countries. From

Generic Framework for Estimating Carbon Footprint of Commuting with Public Transport

Modes

58

1980 to 1997, transportation energy use and

associated GHG emissions increased over 5

percent per year in Asia compared to one

percent growth in greenhouse gases from all

sectors worldwide. Transport situation in most

Indian metropolitan cities is rapidly

deteriorating because of the increasing travel

demand and inadequate public transportation

system and national capital Delhi is no

exception to it.

A large proportion of Delhi’s population

commutes daily for different purposes like

work, education, recreation etc. The mode of

transport chosen for commuting by each

individual depends on his socio-economic

characteristics and the availability of the

modes. Through this study, an attempt has

been made to identify various factors that

affect an individual’s mode choice and then

estimate the carbon footprint due to the

choices made under different scenarios with

main focus on the public transport modes that

are made available to each individual in Delhi.

3.0 OBJECTIVES

The aim of the study is to assess urban

mass transport systems in relation to travel

mode choice for commuting trips. Using

carbon footprint concept to evaluate

sustainability, it is possible to represent and

communicate effectively the issues of

environmental impact and sustainability. The

main objective of the study was to focus on

estimating the carbon footprint due to

commuting focusing on the public transport

modes. The study also estimated the

environmental impact of different travel

options available for commuters. More

importantly, the study evaluated the impact of

Zero carbon modes such as walking, cycling

and rickshaws on carbon footprint of

commuting.

4.0 RESEARCH FRAMEWORK

The framework, as shown in Figure 1, is

designed to be carried out in three phases,

starting from the pre-analysis phase which

includes a general description of the city

transport system, data collection through

passenger interview survey followed by

compilation and preliminary analysis.

Technical part deals with the mode choice

modelling, that acts as a major tool to estimate

the probable division of mode choice between

the two alternatives of public transit. The

mode choice models include socio-economic

variables to account for differences in

individual preferences and level-of-service

variables to measure the relative distributes of

public transit. Analysis part deals with the

estimation of carbon footprint of commuting

during each trip in the study area, i.e. Delhi

using the data collected from the passenger

interview survey and forming different

scenarios for the comparative analysis of the

carbon footprint estimated.

5.0 METHODOLOGY

5.1 Study area

In Delhi there has been a major

improvement in transport infrastructure in

recent years in terms of construction of

flyovers, road widening, new linkages and

operation of metro rail along major travel

corridors. The unprecedented increase in

population, number of vehicles and trips has

put a tremendous pressure on demand of road

infrastructure, but due to resource crunch the

supply has not been able to match the demand.

This has forced the existing network system

function beyond its capacity and has

manifested itself in the form of serious traffic

problems like congestion, delays, safety,

excessive fuel wastage and environmental

pollution.


Modes

59

Figure 1: Conceptual Framework

5.2 Data collection

Passenger interview survey was conducted

at 77 metro stations alongwith adjacent bus

stops for a comparison between the two public

transport modes. The survey areas were

selected with an objective to collect samples

from different socio-economic background and

the areas which are spatially distributed along

the yellow and blue metro line in Delhi as

indicated in figure 2. The yellow and the blue

lines were chosen for the survey as they

represent the North-South and East-West

corridor of the city. The targeted individuals

for the information collection were workers

and students aged above 15. The survey

questionnaire was divided into 4 sections.

These were the socio-economic parameters,

trip characteristics, current travel choice and

their willingness to change to the other mode.

5.3 Modes and Technology Mix

Figure 3 shows the details of the

technology type of various modes in Delhi,

which have also been adopted in this study.

The public modes comprise of Bus and MRTS

which run on CNG and Electricity respectively

whereas personal transport modes comprise of

two-wheeler, motorcycles and cars. The

electricity generation mix in India is as

follows: 70 percent coal, 15 percent

hydroelectricity, 10 percent natural gas, and 5

percent others (mostly petroleum & biomass).


Modes

60

Figure 2: Survey Area for Data Collection

As per the choice set design in the specified

by the respondent for his/her trip and some

from the attached table. To estimate the values

of cost of each OD pair for car, auto rickshaw

and two-wheelers, the data of fuel efficiency,

given in Table 1 and distance traveled is used.

The average operating speeds for metro is

known from the literature along with bus, car,

auto and walking. The speed of cycling is

thrice that of walking and is used for

determination of speed of cycle and cycle

rickshaw, whereas, the speed of two wheeler is

taken as average of car and metro. Information

regarding the fare structure of public bus and

MRTS system is given in table 2. The distance

between origin and destination zones is

estimated using Google Maps®. Similarly, the

distances between the given OD pairs on the

bus network and the metro are also estimated.

Using the information on the distance and the

speed of particular mode (Table 3), the values

for travel time is estimated. Finally, the values

of travel time and travel cost for car, rickshaw,

auto rickshaw, two-wheeler, bus and metro are

used along with other specified variables, to

estimate the utility of each mode.

questionnaire, some variable values were

Table 1: Mode characteristics in Delhi

Vehicle

type

Occup

ancy

Fuel

Efficiency

(Km/lit)

Vehicle

utilization(

Km/year)

Car 2.6 10.9 9,500

Two

wheeler

1.6 44.4 9,000

Three

wheeler

1.8 20 25,000

Bus 60 4.3 70,000

Source: Bose and Srinivasachary (1997)

Figure 3: Modes and Technology Types Mix used in Study


Modes

61

Table 2: Fare structures of the public bus

service and MRTS

Bus MRTS

Up to 4

Km

Rs 5 Minimum Rs 8

4-10 Km Rs

10

Maximum Rs 30

10 Km

onwards

Rs

15

Table 3: Journey Speed for vehicles

Vehicle Speed(Km/hr)

Metro 33*

Two wheelers and

Motorcycles 32

Car 30**

Three wheelers 25

Bus 21***

Cycle 10.5

Cycle rickshaw 7

Walk 3.5

Source: *Gwilliam, K. (2002), **CDP (2006),

***CRRI(2003)

6.0 DATA ANALYSIS

Socio-economic character and the mobility

of the people are directly related to each other.

The more mobile a person is, the wider the

circle of socio-economic interaction that

would be available to them. In turn, both

mobility and socio-economic status influence

the type, frequency and intensity of their

participation in activities. Individual income is

the indicator of socio-economic status. Private

vehicle ownership and public transport

accessibility are the indicators of travel

behaviour.

6.1 Socio-economic and household

characteristics

Socio-economic and household

characteristics such as gender, age, occupation,

household size, household income, vehicle

ownerships, play an important role in the

travel characteristics. Gender and age

distribution of sample are shown in Table 4.

Out of the total responses, majority of the

respondents are working. Majority of workers

are males (80%) and accordingly have higher

percentage in target group. As per age, most of

the individuals belong to the category of 14-30

and 31-50 years of age. Overall, most of the

households belong to the middle income group

having monthly income in the range of Rs.

10,000 to 20,000 per month. The average

household size comes around 4 members per

household.

6.2 Vehicle Ownership

Table 5 shows the vehicle ownership for

sample households. As vehicle ownership

increases, the chances of using public transport

(bus and metro) also decreases. In most of the

cases the vehicle owned by a household is

available to head (male) of the household

while remaining members rely on the public

transport. Household vehicle availability for

different public transport users is given in

Table 6.

Table 4: Socio-economic Characteristics of

Sample Population

Item

Absolute

Values (N)

Relative

Values

(%)

Socio-economic characteristics

No. of individual

observations

4,771

Sex

Male 3,794 79.5%

Female 977 20.5%

Age

14-30 3,014 63.2%

31-50 1,676 35.1%

51+ 81 1.7%

Table 5: Household vehicle ownership

Vehicle Count %

Car 1636 33.3 %

Motor scooter 2204 44.9 %

Bicycle 600 12.2 %

Others 228 4.7%

No vehicle 240 4.9 %


Modes

62

Table 6: Household Vehicle Availability

Public Transport users Household Vehicle Availability

BUS - 1 2 3 4 5 Total

Absolute Values (N) 203 1777 519 122 4 1 2626

Relative Values (%) 7.8 67.7 19.8 4.7 0.2 0.1 100 %

METRO - 1 2 3 4 5 Total

Absolute Values (N) 114 1653 402 73 4 2 2248

Relative Values (%) 5.1 75.5 17.9 3.2 0.2 0.1 100

7.0 CARBON FOOTPRINT

Public transport and non-motorised modes

play a significant role in providing sustainable

transport. There is a lack of comprehensive

detailed study which focuses on the carbon

footprint of different modes of travel. Earlier

work by Bhandari et. al (2010) focuses on the

environmental implications of passenger

mobility in Delhi with focus on the ecological

footprint of commuting. This study focuses on

estimation of carbon footprint based on

complete trip profile from origin to destination

including the access and egress trips and

carbon footprint of different modes of travel.

Figure 4 shows the methodology adopted for

estimating the carbon footprint of travel. The

entire trip from origin to destination is

considered and the mode used in each segment

is considered. The energy and CO2 emissions

from each segment are thus estimated to derive

the final emissions from the trips.

Figure 4: Methodology for calculating Carbon Footprint

Table 7 shows the average emission factors

for Indian vehicles developed by the

Automotive Research Association of India

(ARAI) in 2008. Table 8 gives the average

emission factors of CO2 in tonnes/TJ for each

type of fuel. Using the fuel economy for each

type vehicle the emission factor for CO2 is

derived which is vehicle and fuel specific.

Table 7: Mode Specific emission factors for Indian vehicles

S.No Mode Type CO2 (gms/km)

1. Moped 2 Stroke 33.3

4 Stroke 20.1

2. Two Wheeler 2 Stroke 26.7

4 Stroke 37.9

3. Motor Cycle 4 Stroke 31.8

4. Car Petrol 142.2


Modes

63

S.No Mode Type CO2 (gms/km)

Diesel 156.6

CNG 141.4

5. Three Wheeler CNG 167.7

6. Bus CNG 806.6

Source: ARAI draft report on factor development for Indian vehicles (2007)

Table 8: Emission factor for each pollutant in Ton/TJ

S.

No. Mode Fuel

CO2

(ton/TJ)*

Emission

factor (g/km)

Occupancy

** CO2 (g/pkm)

1 Car Gasoline 68.61 225 2.6 86.53

Diesel 73.33 129 2.6 49.61

CNG 55.82 116 2.6 44.61

2 Two wheeler Gasoline 68.61 55 1.6 34.37

3 Three Wheeler CNG 55.82 13 1.8 7.22

4 Bus CNG 55.82 623 60 10.30

5 Metro Electricity 73.91 1786 1400*** 1.27

Source: *Journal of Urban Planning and Development, Vol 136, No.1, March 1, 2010, pp 89 (2010)

**source of occupancy figures: Bose and Srinivaschary (1997)

***http://www.indianexpress.com/news/metro-starts-shift-to-sixcoach-trains-to-boost-capacity/687516

7.1 CO2 reduction for bus and metro

trips

The pie chart (fig 5) clearly shows that, out

of total data available for all segments- 58% of

trips used metro as main haul mode whereas

42% data were main trip by bus. CO2

emission was then calculated in gram per

passenger for access, egress and main trip, by

multiplying the distance, by respective

emission factor (given in table 8) on the basis

of mode used.

To do this two scenarios were considered

(for access & egress trips) for both bus and

metro trips. In scenario 1 CO2 emission for

both access and egress trips was estimated for

the modes being presently used by both the

metro and bus users. In scenario 2 all the

motorised access and egress trips whose

distance is <=2 km are converted into NMT.

The difference in CO2 of scenario 1 and

scenario 2 for both access and egress shows

the amount of CO2 emission that can be

reduced or saved if we shift to non motorised

trips for distance <=2 km (Table 9).

The reduction in CO2 for access and egress

trips for metro is higher, clearly indicating that

the number of motorised trips/modes being

used to access the metro and to finally reach

the desired destination is higher as compared

to the bus as the main haul trip. This indicates

first and the last mile connectivity for public

transport trips, which are heavily dependent on

the motorised modes in case of metro.

Therefore, by providing better NMT facilities

and infrastructure for NMT modes (walk,

cycle, cycle rickshaw) around metro stations,

carbon footprint of commuting can be reduced

considerably.


Modes

64

Table 9: CO2 reduction for access and egress

trips

Trips

CO2reduction for

A+E trips

(g/passengerkm)

Bus Metro

Segment 3 53.8 70.9

Segment 4 68.4 59.9

Segment 5 23.7 46.7

Segment 6 20.6 25.97

Segment 7 0 0

Total 166.5 203.5

8.0 ANALYSIS OF TRIPS WITH

THREE SEGMENTS

The data is segregated in terms of the

number of trip segments. The total number of

trips finally considered is 4614. Table 10

shows the number of trips of each segment,

with the minimum being 3 segments and the

maximum being 7 segment trips. The three

segment trips were then categorized into 4

categories depending on the mode for access

and egress trips. These categories are NMT-

Metro, NMT-Bus, MT-Metro and MT-Bus.

The number of trips for each of these

categories along with their frequencies is given

below in the table 11. Further, table 12 shows

the percentage of trips where access and egress

trips with motorised and non-motorised

modes. Table shows that 75% of metro trips

have made their access and egress trips by

NMT modes whereas 87% of bus trips have

made their access and egress trips by NMT

modes. This clearly shows that a larger

number of access and egress trips are made by

motorised modes in case of metro.

Table 10: Number of trips for each segment

No of

segments in

a trip

Finally

considered

Data

available

7 33 33

6 66 66

5 368 368

4 1034 1034

3 3113 3373

Total 4614 4874

Table 11: Categories for three segment trips

based on access and egress modes

Category Number of

trips Frequency

NMT-Metro 1311 42

NMT-Bus 1189 38

MT-Metro 441 14

MT-Bus 172 5

Table 12: Mode for Access Egress trips

Total

no of

trips

AE trips

with NMT

modes

AE trips

with MT

modes

Total

metro 1752 74.8 25.2

Total

bus 1361 87.4 12.6

Case-1: As the trip length increases,

proportion of AE distance with respect to total

distance decreases. However mostly this

remains low for NMT bus trips compared to

the NMT Metro trips. This gap is negligible

for trips above 10 km. This is shown in figures

6(A) and 6(B).

Figure 5: Metro and bus users in the sample


Modes

65

Case-2: The share of AE time with respect

to total trip time is shown in figures 7(A) and

7(B). As the trip length increases, proportion

of AE time with respect to total time

decreases. However this always remains low

for NMT bus trips compared to the NMT

Metro trips. The gap between NMT Metro and

NMT bus trips remains nearly the constant.

This indicates that it is easy to get a bus using

NMT modes as feeder than to get a metro.

Case-3: There is no clear relationship

observed for the cost factor as shown in

figures 8(A) and 8(B). This includes the cost

attached with the A&E trips which have been

made by NMT modes. This highlights the need

of further work. This study would be further

extended by separate analysis for each type of

NMT modes used for making these A&E trips.

Figure 6(A): Share of Access and egress trips

distance in total distance by NMT modes Figure 6(B): Share of Access and egress trips

distance in total distance by MT modes

Figure 7(A): Share of Access and egress trips

time in total time by NMT modes Figure 7(B): Share of Access and egress trips

time in total time by MT modes.

Figure 8(A): Share of Access and egress trips time in

total cost by NMT modes.

Figure 8(B): Share of Access and egress trips

distance in total cost by MT modes


Modes

66

9.0 CONCLUSIONS

The transport sector has a potential impact

on the GHG emissions and ultimately results

in Climate Change. This sector contributes

about 14% to the global GHG emissions.

Thus, curbing these emissions has become an

area of concern for transport planners,

engineers and environmentalists. Only the

aggressive strategies can slow down the pace

of increase in GHG emissions. This study

focused on two different aspects of

commuting, one the mode choice for each

individual from a given set of options and the

other carbon footprint accrued due to their

commuting choices. The study results showed

that 28% of the commuters were willing to

shift to metro and the carbon emissions

occurring because of bus users was way higher

than those of metro users. Thus, the shift of

commuters from the carbon intensive mode i.e.

bus to metro would further help in reducing

the impact of commuting in terms of carbon

emissions. But reducing the CO2 emissions by

upgrading commuting to “greener”

transportation modes will require an array of

coordinated, progressive transportation

policies, supplemented by public outreach

campaigns like, educating people on carbon

impacts of commuting by personalized or road

based public transport modes as well as

benefits of using the Zero- polluting

commuting options. The trips performed for

work and education purpose are consistent,

predictable, performed alone and generally

within manageable distances, therefore, there

is a scope to alter commuting trip by way of

switching to a different mode as compared to

other less predictable trips.

Moreover, the repetitive nature of trips

implies that huge benefits in terms of carbon

footprints can be accrued from seemingly

small changes.

REFERENCES 1. Automotive Research Association of India

(ARAI) (2007). Emission factor

development for Indian vehicles.

2. Bhandari, K., Shukla, A., Gangopadhyay,

S., Hayashi, Y.(2010). Environmental

implications of passenger mobility in

Delhi: Energy consumption, CO2 emissions

and ecological footprint of commuting, 3rd

international conference of Transport

Science and Technology Congress, Apr 4-

7, New Delhi.

3. Bhandari, K. And Hayashi, Y. (2011).

Mass Rapid Transit and its Impact

Assessment: Case of Delhi, Economy,

Equity and Environment, LAP LAMBERT

Academic Publishing.

4. Bose, R. K. and Srinivasachary, V.(1997)

Policies to reduce energy use and

environmental emissions in transport sector

A case of Delhi city, Energy Policy, Vol.

25(14-15), pp.1137-1150.

5. Central Road Research Institute (CRRI)

(2003), Urban Road Traffic and Air

pollution.

6. Gwilliam, K. (2002). Cities on the

move:Urban transport strategy review,

World Bank, Washington D.C.

7. Han, J, Bhandari, K. And Yoshitsugu

Hayashi. (2010) Assessment of Policies

towards an environmentally friendly urban

transport system: Case study of Delhi,

India, Journal of Urban Planning and

Development, Vol 136, No.1, March 1, pp

89

8. IEA (2007). CO2 emissions from fuel

combustion, 1971-2005: 2007 edition, IEA

Paris

9. http://web.worldbank.org/WBSITE/EXTE

RNAL/TOPICS/EXTTRANSPORT/0,,cont

entMDK:21517582~menuPK:337124~page

PK:148956~piPK:216618~theSitePK:3371

16,00.html last accessed April 17, 2012

10. http://oica.net/category/climate-change-

and-co2/ last accessed on August 28, 2011.

URBAN TRANSPORT

JOURNAL


* M.Tech Student, Academy of Scientific and Innovative Research (AcSIR), CSIR-Central Road Research

Institute New Delhi-110025, [email protected]

** Senior Scientist, Transportation Planning Division, CSIR-Central Road Research Institute, New Delhi

10025, [email protected]

MODELING MODE CHOICE BEHAVIOUR AND ESTIMATING

VALUE OF TRAVEL TIME OF COMMUTERS IN DELHI

Minal* and Ch.Ravi Sekhar**

Abstract: Dealing with the present bottlenecks as well as creating long lasting and sustainable transport systems

has been the greatest challenge of urban transport planning. Calibrating the present need and forecasting the

future demand is the underlying agenda of travel demand forecasting. Mode choice forms an integral part of this

process as it gives a complete insight to the mode choice preferences of the commuters validating the

introduction of new transport systems to existing ones. This study aims at modelling the mode choice of

commuters in Delhi. For this discrete Multinomial logit has been considered and mode choice analysis has been

carried out. An extensive household survey has been carried out and disaggregate data was collected in various

localities of Delhi. Thirteen explanatory variables were considered which includes household information,

personal information and trip information for modeling mode choice behavior. Value of travel time has been

quantified separately for motorized and non motorized mode of commuters. The value of in-vehicle travel time

estimated for motorized vehicle is found to be 95 Indian Rupees (₹) per hour. The value of total travel time

estimated for non-Motorized vehicle is found to be ₹451 per hour.

Keywords: Mode Choice, Delhi, household survey, multimodal logit, value of travel time

1.0 INTRODUCTION

Transportation community is bound to face

challenges that are both dynamic in nature and

futuristic in its application perspective.

Amongst the various confluences in

Transportation system, congestion is by far the

most common and difficult factor to

overcome. Congestion drastically affects the

level of service of the transport system leading

to consequences like delay, accidents which

lead to huge economic loss every year. To

alleviate the situation studying the travel

behaviour and choice of commuter is useful.

The ultimate interest lies in being able to

predict the decision making behavior of the

commuters while taking under consideration

the attributes of different modes like cost,

safety, convenience and travel time. Mode

Choice problem has been approached by

transportation planners in many different

ways. In a broad way all these approaches can

be classified into two categories namely

discrete choice models and non-discrete choice

models. Discrete choice models primarily

include Multinomial Probit model,

Multinomial Logit model (MNL) and Nested

Logit (NL) model. Non-discrete choice models

include regression approach, cross

classification tables and diversion curves. The

objective behind mode choice model is to

effectively manage the demand and be able to

provide for these demands by making changes

in the existing system.

In Delhi major modes of transport are

private cars, two wheelers, bus, metro and auto

rickshaw (three wheeler) and bicycle. Delhi

with a population of 16.8 millions (Census,

2011) is under constant need of expansion of

existing transport facilities. Attracting the

users of private modes to mass transport

modes like bus and metro seems to be a

solution but is not very feasible given the

comfort factor of mass transport facilities.The

objective of the present study was to develop a

mode choice model for commuter of Delhi by

considering most widely used Multinomial

Logit (MNL) model. Data collection of the

disaggregate data was done through an

extensive household survey in Delhi to

account different strata of population with

Modeling Mode Choice Behaviour and Estimating Value of Travel Time of Commuters in Delhi

68

different socioeconomic backdrops, age and

gender. The relative influence of various

parameters associated with different modes

also been estimated and it’s interpretation has

been elaborated. This will help to policy

makers or transportation engineers/planners

for further improvements and amendments of

the transit facilities in Delhi. The value of

travel time has also been estimated in this

study to quantify the economic measure of the

commuters travel time.


Mode choice model initially proposed by

Adam (1959) to investigate factors influencing

mass transit and automobile travel in urban

areas. Mode choice models are two folds

aggregate and disaggregate models. Aggregate

models are based on zonal or inter zonal

information and disaggregate models are based

on household or individual data also referred

to as micro data. Disaggregate models

associated advantages over aggregate models,

has led to the widespread use of disaggregate

discrete choice methods in travel demand

modeling, destination choice (Koppleman and

Bhat,2006), route choice (Gliebe et. al, 1998),

air travel choices (Proussaloglou et. al ,1999)

activity analysis (Wen et. al, 1999) and auto

ownership, brand and model choice (Bhat et.

al, 1997).

Bhat (1995) developed a heteroscedastic

extreme value model of intercity mode choice

in which estimation of the ridership share on a

proposed new intercity travel service was

done. Identification of the modes from which

existing intercity travellers will be diverted to

the new or upgraded service was performed.

Five different models in the study were used

there were multinomial logit model, three

possible nested logit models, and the

heteroscedastic extreme value model. The

resulting heteroscedastic extreme value model

has a number of advantages over other

commonly used discrete choice models. Al

Ahmadi (2006) developed intercity mode

choice models for Saudi Arabia. In this study

he considered Multinomial logit model for the

model development. Data collection was done

through “revealed preference” surveys. The

results indicated that in-vehicle travel time, out

of pocket cost, number of family members

travelling together, monthly income, travel

distance, nationality of traveller, and number

of cars owned by family played the major role

in decision related to intercity mode choice.

Khan (2007) estimated various nested logit

models for different trip length and trip

purpose using data from stated preference (SP)

survey. A unique set of access modes for bus

on bus way was generating containing

hypothetical modes such as secure park and

ride facilities and kiss and ride drop-off zones.

He found that the travel behaviour forecasted

for regional trip makers is considering

different from that for local trip makers.

In India, most of the modeling approaches

are oriented towards the use of economic

theory of Utility maximization. Many

researchers have developed mode choice

models based on principles of utility

maximization (Chari 1978). Whereas disutility

of minimization employed by Rao (1988).

Parida (1994) has employed stated and

revealed preference approaches for modeling

home based work trips in Delhi. Subbarao et.al

(1997) developed access mode choice model

using ANN and compared the results with

conventional Multinomial Logit model

(MNL). Ravi Sekhar (1999) developed mode

choice model by using ANN and MNL for

Delhi data. In this study, data has been

classified based on vehicle ownership. Ravi

Sekhar et.al (2009) studied on applications of

Neural Networks in mode choice modelling

for second order metropolitan cities of India

for this they have considered second order

cities travel behavior data in the cities of

Visakhapatnam and Nagpur.Ashalata et al.

(2013) attempted a revealed preference study

of mode choice for Thiruvanathpuram city

using Multinomial logistic regression. The

major modes included in the study were car,

two wheeler and bus. The analysis highlighted


69

the fact that preference to car increases with

age while the preference to use of two wheeler

decreases in comparison to bus.

3.0 STDY AREA AND DATA

COLLECTION

3.1 Study Area

In this study commuter in Delhi was

considered. Delhi has a population density of

11,297 per square km. Mobilising such dense

population in a metropolitan which houses

multiple offices, industries and manufacturing

units is a marathon task. The city suffers with

the problem of Congestion, Road fatalities and

high levels of Pollution. The use of private

vehicles (Drive alone cars and Two wheelers)

is very familiar and plays a devious role in

choking the networks of the city in peak and

off peak hours. The city has been mainly

divided four zones namely East, West, North

and South Delhi, Figure 1 represents the study

area. A total sample size of 5000 households

interviews were collected. The numbers in the

popup represents the number of samples

collected from different zones. The survey

questionnaire and method of data collection

was briefly discussed in the subsequent

sections.

Figure 1: Study Area and Sample size of Data

Collection from each zone

3.2 Questionnaire Design

The questionnaire was comprised of four

sections namely household information,

person information, trip Information and

vehicle Information. Household Information

includes household size, number of earned

persons in household, household income,

monthly household travel expenditure and

dwelling unit type. Personal information such

as age, gender, education level, occupation and

possession of driving license parameters was

considered. Travel Distance (Home to work),

travel time, access time, waiting time, transfer

time, Parking time, egress time, travel cost and

preferred mode of travel was pursued under

trio information. Vehicle ownership, number

and type of vehicles owned were captured

considered under Vehicle information.

Ranking type questions were also considered

for modal serviceability attributes like cost,

security, hygiene, privacy, travel time

reliability, waiting time were to be ranked for

public mode.

3.3 Travel Behavior Data Collection

In the present study, household interview

survey was conducted by CSIR- Central Road

Research Institute through predesigned

questionnaire. In all 5000 house hold sample

were collected in Delhi. Stratified random

sample was considered. In the first stage of

sampling, blocks or clusters of colonies was

recognized. In the second stage of sampling

particular Households (HH) were identified

and household members were interviewed.

This type of sampling avoids any sort of

biasness in the data collection procedure. The

interviewers visited the pre- identified pockets

and interviewed the household member. A

sample size of 3000 survey responses was

collected from South Delhi and 2000 samples

were collected from the North, East and West

Delhi.

From the data, it was observed that the

largest commuter share comes from the age

group of 31 to 50 years with approximately

20%, 12%, 15% of them using drive alone car,

two wheelers and bus respectively. The female

commuters’ most preferred mode of travel is

drive alone car which definitely provides

higher security and privacy to women. The

house hold income of less than ₹10000 per


70

month mostly uses two wheelers, buses and

walk. The variables considered to influence

the mode choice behaviour and available

choice of mode are (as per the survey data)

listed in Table 1. Traditionally such

explanatory variables include

household/personal, socio-demographic and

trip information.

Table 1: Explanatory variables Choice Variables considered for mode choice analysis

Description of Model Input Variable Variable Type

House hold Information

Household size Continuous

Number of vehicles in house hold Continuous

Household income (Indian Rupees) Continuous

Personal Information

Age of traveler in years Continuous

Gender of traveler 1:Male; 0:Female

Education Level 1.Elementary; 2.Intermediate; 3:Diploma;

4:Graduate; 5: Postgraduate; 6:Doctrate;

7:Post Doctorate

Type of employment 1:Student; 2:Govt.employee;3:PrivateSector

Employee;4:Business Owner;5:Other

Possession of Driver’s License 1: Yes; 0:No

Trip Information

Trip Purpose 1: Work trip; 0: Non Work Trip

Serviceability provided by Mode in use Continuous

In Vehicle Travel time for MV (Minutes) Continuous

Out of Vehicle Travel time for MV (Minutes) Continuous

Travel cost ( Indian Rupees) Continuous

Total Travel Time for NMV ( Minutes) Continuous

Available Mode Choice Discrete

Drive Alone Car (Private mode)

Carpool (Shared mode)

Two Wheeler (Private mode)

Bus (Public mode)

Metro (Public mode)

Auto Rickshaw (IPT mode)

Bicycle (Personal/ Non motorized mode)

Walk (Non motorized mode)


71

4.0 MODE CHOICE ANALYSIS

4.1 Development of Multinomial Logit

Model

Logit models are mainly three types,

namely Multinomial Logit Model (MNL),

Conditional logit model (CL) and Mixed Logit

model (MXL). Logit models depending on

whether the data are chooser-specific or

choice-specific. MNL model has chooser

specific data where coefficients vary over the

choices. CL model has choice-specific data

where the coefficients are equal for all

choices. MXL model involves both types of

data and coefficients. MNL model is widely

used disaggregate mode choice model, it

estimates the proportion of trip makers who

choose available mode types based on given

conditions or utility criteria. MNL model is

often used to compare with other techniques,

due to its ability in analyzing the trip maker

behavior (Hensher et. al,2000). MNL model

considered in this study to model choice

behiour commuters in Delhi. The

mathematical framework of logit models is

based on the theory of utility maximization

(Ben-Akiva and Lerman, 1985). Probability of

an individual "i" selecting a mode "n", out of

"M" number of total available modes, is given

in equation (1).

𝑷𝒊𝒏

=𝒆𝑽𝒊𝒏

∑ 𝒆𝑽𝒊𝒎𝑴𝒎=𝟏

𝐸𝑞𝑛. (1)

Where, Vin is the utility function of mode

"n" for individual "i" , Vim is utility function

of any mode "m" in the choice set for an

individual "i". Pin is the probability of

individual "i" selecting mode "n". M is the

total number of available travelling modes in

the choice set for individual "i". However, the

Logit model has certain drawbacks like

requirement of large sample size and

restriction on dependent variable to be of

discrete dataset.

4.2 Results and Discussion of MNL

Model

In this study four different mode choice

models were developed to evaluate the

behaviour of commuter to choose the

particular mode.

Base Model (M1): Initially, base model

(M1) was developed, this model consisted of

basic travel parameters like travel time which

includes In Vehicle Travel Time(IVTT), Out

of Vehicle Travel Time(OVTT) and House

hold income as an Alternative Specific

Constant (ASC). The inclusion of House hold

income in the model will reflect the biases that

by each commuter will have for mode

selection with respect to change in Income.

Walk was taken as the reference mode.

Model 2 (M2): This model consists of

incremental improvements on the specification

of base model (M1). Improvement in model

was explored by addition and interaction of

various travel parameters. The possession of

Driving License (DL) was included as a

dummy variable. Other variables like age,

gender and education level was also

incorporated in the model specification.

Number of workers in household, trip purpose

and modal service level expected by the

commuters was also included. Model M2

shows great improvements in its log likelihood

value. The value obtained at convergence

shows a value of -2926.61 as compared to

Model M1 value of -4388.52. Also by log

likelihood ratio test M2 rejects M1

significantly.

Model 3 (M 3): To get a better insight into

the travel behaviour of motorized and non

motorized mode commuters, the travel time

was split into two types in model 3 The first

part was the total travel time for Non

Motorized mode (NMV) i.e. Bicycle and Walk

and the second part was that of IVTT and

OVTT for Motorized vehicles (MV).

The estimated MNL model coefficients for

each model were presented in Table 2 with


72

changes in their utility function specifications.

The influence of explanatory parameter in

each model was evaluated through t-static

value, which is the ratio of the model

coefficient to standard error estimate. This is

presented in the brackets of Table 2. The t-

static values higher than ± 1.95 at 5% level of

significance have significant contribution in

the model development. The detailed result

and discussion are described in next section.

Model 4 (M4): In Model M4 interaction

amongst variables were tried for further

betterment of model. OVTT of motorized

vehicles was interacted with the logarithm of

distance travelled. Trip length is a trip

parameter that will bring forth the sensitivity

of commuters with respect to travel distance.

The model shows a slight improvement on

employing this variable interaction. Like hood

ratio test was performed for validating the

overall goodness of fit of model. The

likelihood ratio test statistic is twice the

difference in log-likelihoods of the two models

under consideration (-2(LLbase model –

LLestimated model). For example LL ratio

tests of Model M2, Model M3, Model M4

w.r.t. Base Model M1. The log like hood value

has improved for Model3 significantly after

this breakup of travel time component. Model

M4 was found to be the most significant one.

So, the coefficients used in model 4 are more

appropriate and further considered for

evaluating Value of travel time. From Akaike

Information Criteria (AIC) values it was

observed that Model 4 has the least AIC value

compared to rest of the models indicates that

this model is the one with the least divergence

from truth (but few parameters). The DA car

mode greatly dominates the work trips,

followed by carpool and two wheelers. Bus,

auto rickshaw and bicycle have a negative

influence on the work trip makers. Metro has a

relatively little although positive attraction for

work trips. In MV the IVTT and OVTT has

almost the same influence on travel. But the

value obtained for IVTT of MV is more

significant than the OVTT. Travel cost also

has a significant and negative influence on

travel where increase in cost is observed as

undesirable. But the TT for NMV has a greater

(negative) influence on the walk and bicycle

modes. The TT_NMV is also highly

significant and cannot be ignored.

It was observed from the coefficients of the

household income that the utility of drive

alone car and carpool increases with

increasing income. The lower income group

makes use of the transit and para transit

facilities more than other modes. Travelers

with higher Education level the utility of bus,

two wheelers, auto rickshaw and bicycle is

very low. The older age group is inclined

towards walking or shared car drive. The

younger age group prefers use of two

wheelers, metro and auto rickshaws while

drive alone car is most popular with the

middle age group commuters. Gender wise,

two wheelers, carpool, bus and bicycles is

preferred more by male travelers while female

counterparts prefer use of metro, auto

rickshaws and drive alone cars. Household

with more number of working persons are

more inclined towards carpooling and use of

bus. For work trips drive alone car is the most

popular mode of travel while bicycle is the

least preferred one. The household that have a

higher budget for traveling prefer use of car

(alone and shared), Two wheelers and metro.


73

Table 2: Estimation of Model Coefficients of Various MNL models

Modal Parameters Base Model 1

(M1) Model 2(M2) Model 3 (M3) Model4 (M4)

IVTT .03230 (1.65) -.05089 (2.42)

OVTT .00640 (0.55) -.03371 (2.67)

TT_NMV -.17529 (-6.17) -.188 (-7.74)

IVTT_MV -.03226 (-2.09) -.039 (-2.85)

OVTT_MV -.00061 (-0.08)

OVTT_MV/log(Distance) -.036 (-1.52)

Travel Cost -.09338 (11.62) -.06497 (7.38) -.02464 (3.86) -.025 (3.97)

Alternative Specific

Constant (ASC)

Drive Alone Car -1.169 (0.24) -1.173 (-3.23) -2.896 (-3.43) -2.47 (-3.18)

Car Pool -6.050 (23.75) -9.812 (-7.9) -12.46 (-8.63) -12.25 (-8.55)

Two-wheeler -1.063 (9.45) 3.335 (-0.12) .172 (0.02) 0.46 (0.13)

Bus -3.760 (17.64) 3.693 (1.9) 1.026 (1.28) 1.51 (1.62)

Metro -6.334 (23.06) .221 (-.99) -1.045 (-1.07) -0.61 (-0.85)

Auto-Rickshaw -1.221 (22.45) 2.576 (0.63) 1.389 (0.74) 1.99 (0.92)

Bicycle 0.0 (1.92) 12.66 (2.98) 8.543 (3.06) 9.33 (3.09)

Modal Serviceability

Drive Alone Car .0018 (.12) .001 (0.26) 0.004 (0.30)

Car Pool .1030 (3.51) .103 (3.81) 0.104 (3.75)

Two-wheeler .0191 (1.28) .019 (1.58) 0.022 (1.56)

Bus .0669 (4.80) .066 (5.80) 0.065 (5.86)

Metro .0191 (0.92) .019 (0.94) 0.017 (0.94)

Auto-Rickshaw .0342 (1.64) .034 (1.84) 0.046 (1.89)

Bicycle .0173 (1.42) .017 (1.59) 0.055 (1.62)

HH Income

Drive Alone Car -.3340 (0.23) .658 (-2.52) -.5077 (-2.83) -0.0647 (-2.84)

Car Pool .8499 (23.75) .650 (0.43) 0.0 0001 (0.23) 0.0059 (0.19)

Two-wheeler -.1012 (6.37) -.876 (-3.63) -.8682 (-4.03) -0.0867 (-4.06)

Bus .1401 (1.27) -.361 (-2.33) -.4467 (-2.65) -0.0362 (-2.70)

Metro -.452 (1.76) -.890 (-2.89) -.9763 (-3.09) -0.0895 (-3.11)

Auto-Rickshaw .6474 (22.06) .126 (0.12) 0.0 (0.55) 0.0137 (0.52)

Bicycle -.0001 (3.65) -.415 (-1.85) -.4379 (-1.95) -0.9706 (-1.98)

Gender

Drive Alone Car 8617 (2.63) .8617 (2.99) 0.8723 (2.99)

Car Pool -1.328 (2.42) 1.328 (2.67) 1.3194 (2.62)

Two-wheeler -1.581 (-4.53) 1.581 (-4.03) 1.5859 (5.91)

Bus 1.333 (-2.45) 1.333 (-2.65) 1.3098 (7.21)

Metro 6334 (-2.90) .6334 (-3.09) 0.5941 (1.82)

Auto-Rickshaw 5931 (0.21) .5931 (0.55) 0.6813 (1.60)

Bicycle 2.063 (-1.35) 2.063 (-1.95) 2.2130 (1.96)

Trip purpose

Drive Alone Car -.3.524 (9.31) 3.514 (9.28) 3.5098 (9.27)

Car Pool 1.267 (2.01) 1.267 (1.88) 1.2440 (1.83)

Two-wheeler .8445 (2.89) .8449 (2.68) 0.8567 (2.64)

Bus --.6452 (-2.93) -.6098 (-2.53) -0.6467 (-2.55)


74

Modal Parameters Base Model 1

(M1) Model 2(M2) Model 3 (M3) Model4 (M4)

Metro .54432 (0.91) .4443 (0.79) 0.3777 (0.77)

Auto-Rickshaw -1.778 (-2.01) -1.678 (-1.97) -1.3179 (-1.95)

Bicycle 1.767 (-1.62) -1.797 (-1.20) -1.6497 (-1.21)

LL at Zero -9237.34 -9237.34 -9237.34 -9237.34

LL at Constants -5478.19 -5478.19 -5478.19 -5478.19

LL at Convergence -4388.52 -2926.61 -2902.81 -2900.54

AIC 1.801 1.229 1.223 1.219

Likelihood Ratio Test NA 2923.82 2971.42 2975.96

Prediction Accuracy 52% 70% 72% 74%

5.0 ESTIMATION OF VALUE OF

TIME

Value of travel time (VOT) plays a crucial

role in the cost benefit analysis in transport

planning process. It quantifies the importance

of time with respect to cost which is employed

in economic evaluation of travel time saving.

The most standard procedure suggests Value

of Time as a trade off ratio between coefficient

of travel time and travel cost. VOT is equal to

the ratio between the derivative of utility with

respect to time and the derivative of utility

with respect to cost, mathematically expressed

in equation (2)

VoT

=

∂Vi

∂Timei∂Vi

∂Costi

⁄ Eqn. (2)

For utility functions that are interacted with

other variables the VOT formulation changes

slightly. This is expressed mathematically in

equation (3) & (4).

VoTIVTT =

∂Vi

∂TimeiLog(Distance)

⁄

∂Vi

∂Cost

Eqn(3)

VoTOVTT

=

∂Vi

∂TimeiLog(Distance) + βIVTT

⁄

∂Vi

∂Cost

Eqn(4)

Where, Log (Distance) is the Log (Trip

distance) and βIVTT is the coefficient of IVTT

obtained by model estimation.

Value of In Vehicle Travel time as computed

using the MNL model (w.r.t. Trip length) is

presented in Table 3 row one. The value of

travel time for In vehicle travel time for MV is

obtained as 95(₹ /Hr.). Value of Out of

Vehicle travel time is highest for shorter trip

and decreases as the trip length increases

which imply that commuters are more

sensitive to waiting time, parking time, access

and egress time for shorter trips. When the trip

length increases travelers are concerned more

with the In Vehicle travel time and less with

Out of Vehicle travel time. Also the travelers

in non motorized modes are expected to be

more sensitive to travel times as walking and

bicycling are physically more demanding than

the motorized modes (Koppleman and Bhat,

2006). The value of Travel time for non

motorized modes is in accordance to our

expectation and are very high (₹ 451/hr)

compared to value of travel times of motorized

modes.

But in real case scenario not every

individual values time equally. It is a matter of

taste variations and personal preferences. To

account for this heterogeneity, VOT should be

quantified for different segments of the

population under consideration based upon the

attributes of involved society and the

characteristics of the trip. Segmentation based

on a) Gender and b) Trip Purpose has been

done to investigate how it affects the VOT for


75

different situations. The Table 3 represents the

VOT for the above mentioned stratifications.

Heterogeneity in the population is clearly

indicated by the VOT calculated for different

segments.

The gender classification of population

depicts results that exhibit large heterogeneity

in perception of travel time among male and

female travelers. Male travelers are more

sensitive towards the travel time. They

perceive both the IVTT and OVTT for the trip

made more critically than their female

counterparts. But what is more pertinent is the

high value of non motorized travel time.

Female commuters perceive this time much

more critically (approximately 100 times

IVTT of MV). It subtly brings forth the safety

aspect of NM modes of travel where walking

and bicycle has been considered. The social

security of female commuters plays a major

factor in this aspect as walking to work from

home (and back) has very high chances of eve

teasing and other unwanted hassles that female

commuters face every day.

Comparing the Work and Non work trips of

travel; work trips have a very high value of

travel times as expected. A work trip IVTT is

valued 5 times more than IVTT of non work

trips. During work trips the Out of Vehicle

travel time spent for waiting, transfer and

parking are all very crucial and the commuters

seem to be very sensitive towards it. It

implicitly suggests that the service level of

operating modes affects the travelers more

than the explicit factors in making their mode

choice decision. Even the NM modes (i.e

Walk and bicycle) place a very high value of

their travel time (443 ₹ /Hr). This in fact

indicates the plausible cause why the NMV

have become so unpopular in making trips.

Non work trips show great variation with

IVTT being valued at just 24 ₹ /Hr while

OVTT is valued almost 3 times higher (73 ₹

/Hr.) This result asserts that for non work trips,

commuters are very critical about comfort and

convenience and every minute spend in

waiting, walking or parking is deemed as very

crucial.

Table 3 Value of Time for different Segments based on Trip length

5 Km trip 10 Km trip 15 Km trip

Value

of

IVTT

of MV

Value of

OVTT

of MV

Value

of TT

of

NMV

Value

of IVTT

of MV

Value of

OVTT

of MV

Value

of TT

of

NMV

Value

of IVTT

of MV

Value of

OVTT

of MV

Value

of TT

of

NMV

MNL Model

(Pooled

Model)

₹ 95/hr ₹ 122/hr ₹

451/hr ₹ 95/hr ₹ 85/hr

₹

451/hr ₹ 95/hr ₹ 73/hr

₹

451/hr

Segmentation based on: Gender

Male ₹

127/hr

₹ 156/hr ₹

464/hr

₹ 127/hr ₹ 110/hr ₹

464/hr

₹ 127/hr ₹ 92/hr ₹

464/hr

Female ₹ 3/hr ₹ 35/hr ₹

366/hr

₹ 3/hr ₹ 24/hr ₹

366/hr

₹ 3/hr ₹ 20/hr ₹

366/hr

Segmentation based on: Trip Purpose

Work trip ₹

122/hr

₹ 132/hr ₹

443/hr

₹ 122/hr ₹ 92/hr ₹

443/hr

₹ 122/hr ₹ 78/hr ₹

443/hr

Non work

trip ₹ 24/hr ₹ 73/hr ₹

222/hr ₹ 24/hr ₹ 51/hr ₹

222/hr ₹ 24/hr ₹ 43/hr ₹

222/hr

6.0 SIGNIFICANT FINDINGS

This study focused on the mode choice

analysis of Delhi which is subjected to heavy

congestion and pollution due to high number

of private vehicles plying on the roads. The

data collection was done through a household

mode choice models were developed and

various explanatory variables were


76

incorporated to improve the goodness survey

in Delhi. A large Household survey sample of

5000 responses was collected and discrete

disaggregate Multinomial logit model was

considered for carrying out mode choice

analysis. Four different Information Criterion

value it can be concluded that M4 has better

predictability in modelling mode choice

behaviour. Therefore the coefficients obtained

from Model M4 were employed for

investigating and estimating the value of travel

time. the fit and interpretability of the MNL

model. Model M4, from both log likelihood

value and Akaike

Travel time is a valuable nonrenewable

resource and people devote a great deal of

their time in travelling and thus it forms a

strong basis for evaluation of any transport

system. Travel time savings is often the

principal benefit of a transportation project.

The findings of this study be used in

congestion relief projects are justified

primarily by the reduction in travel time they

will bring about. Travel time savings can also

lead to reductions in vehicle operating costs.

Thus the VOT estimated in the study can

prove useful in case when new policies and

infrastructure are appended to the

transportation system. The VOT will render

useful in the cost benefit analysis.

Further, the following are the significant

conclusions drawn from the results.

The coefficients obtained in the model4

(M4) indicate that affluent and higher

income household prefers Drive alone Car

and Carpool as their major mode choice

decision which is also indicative of a lavish

lifestyle.

The interaction of variables in model 4

(M4) gives a better result with increases

predictability of 74% compared to M1, M2

and M3 which have a prediction accuracy

of 52%, 70% and 72% respectively.

The strata of society with higher level of

education are least inclined towards use of

Public and Inter Para Transit facilities.

Female commuters prefer Drive Alone car,

Metro and Auto Rickshaw than other

available modes.

It is found that Value of Travel time for

NMV is 5 times the Value of In Vehicle

Travel Time for MV which is expected as

each minute spend in walking and cycling

is more demanding than the same time

spend in travelling through a motorized

vehicle

REFERENCES 1. Adams, W.T. (1959), Factors Influencing

Mass Transit and Automobile Travel in

Urban Areas, Public Transport, (30), pp

256-260.

2. Ashalata, R., MAnju, V.S. and Zacharia, A.

B.(2013), “ Mode Choice Behaviour of

Commuters in Thiruvananthapuram City ”,

Journal of Transportation Engineering,

American Society of Civil Engineers,

Volume 139, Issue 5, pg 494-502.

3. Abdel-Aty, M. and Abdelwahab, H. (2001),

“Calibration of nested mode choice model

for Florida”, Final research report,

University of central Florida.

4. Al-Ahmadi, H.M. (2006), “Development of

Intercity Mode Choice Models for Saudi

Arabia”, JKAU: Eng.Sci, Vol. 17 No. 1 ,

pp. 3-12.

5. Ben-Akiva, M. E. and Lerman, S. R.,

(1985), “Discrete Choice Analysis: Theory

and Application to Travel Demand”, The

MIT Press, Cambridge, Massachusetts, the

USA.

6. Bhat, C.R., and R. Sardesai (2006), "The

Impact of Stop-Making and Travel Time

Reliability on Commute Mode

Choice," Transportation Research Part B,

Vol. 40, No. 9, pp. 709-730

7. Bhat, C.R. (1995), "A Heteroscedastic

Extreme Value Model of Intercity Mode

Choice", Transportation Research Part B,

Vol. 29, No. 6, pp. 471-483.

8. Bhat, C. R. and Pulugurta, V. (1998). A

Comparison of Two Alternative Behavioral

Mechanisms for Car Ownership Decisions.


77

Transportation Research Part B, 32(1), 61-

75.

9. Gebeyehu, M. and Takano S,(2007),

“Diagnostic Evaluation of Public

Transportation Mode Choice in Addis

Ababa”, Journal of Public Transportation,

Vol. 10, No. 4

10. Gliebe, J.P., F.S. Koppelman and A.

Ziliaskopoulos (1998) Route choice using a

paired combinatorial logit model, prepared

for presentation at the 78th meeting of the

Transportation Research

Board,Washington, D.C., January 1999.

11. Hensher, D. A. and T. Ton (2000). TRESIS:

A transportation, land use and

environmental strategy impact simulator

for urban areas. Transportation 29(4):

439-457.

12. Khan,O. (2007), “Modelling Passenger

Mode Choice Behavior Using Computer

Aided Stated Preference Data”, P.H.D.

thesis, Queensland University of

technology

13. McDonald, N.C.(2008). “Children’s mode

choice for the school trip: the role of

distance and school location in walking to

school”. Transportation Vol. 35 No.2,

2008, pp. 23-35.

14. McFadden, D. (1978), Modelling the

choice of residential location, in A.

Karlquist, ed., ‘Spatial Interaction Theory

and Planning Models’, North Holland,

Amsterdam, pp. 75–96

15. Mukala, P.K and Chunchu, M.(2011),

“Mode choice modelling for intercity

transportation in India:A case of Guwahati

to five metro cities”, International Journal

of Earth Sciences and Engineering. Volume

04, No 06 SPL, pp. 364-374.

16. Parida,M.,(1994) Mode Choice Analyis

Based on Stated and revealed Preferences

for Home Based Work Trips in Delhi, PHD

Thesis Department of Civil Engineering,

IIT Roorkee, Roorkee,India.

17. Proussaloglou, K., F. S. Koppelman, 1999.

The Choice of Air Carrier, Flight, and Fare

Class.Journal of Air Transport

Management 5 (4), 193-201.

18. Provisional Population Totals Paper 1 of

2011 : NCT of Delhi, Chapter-2, Data and

Major Trends,pp 45.

19. Ravi Sekhar Ch. (1999) Mode choice

analysis using Neural Network,

Department of Civil Engineering, Master

Dissertation, IIT Roorkee, Roorkee, India.

20. Ravi SekharCh.,Madhu,E, Durai,B.K and

Gangopadyayay. S Applications of Neural

Networks in Mode Choice Modelling for

Second Order Metropolitan Cities of India,

Proceedings of the Eastern Asia Society for

Transportation Studies, Vol.7, 2009

21. Subba Rao, P. V., Dhingra, S. L., Sikdar, P.

K. and Krishna Rao, K. V. (1997) Access

mode choice analysis using artificial neural

networks, Proceedings of the Conference

on Trends and Techniques of

Transportation, REC Warangal, India.

1997, 81-96.

22. Sidharthan,R., Bhat,C.,Pendyala,R. and

Goulias, K.,(2011) “A Model Of Children’s

School Travel Mode Choice Behavior

Accounting For Spatial And Social

Interaction Effects”, Transportation

Research Record: Journal of the

Transportation Research Board, vol.

2213,pp78-86.

23. Wen, C. and F.S. Koppelman (1999) An

Integrated Model System of Stop

Generation and Tour Formation for the

Analysis of Activity and Travel Patterns,

forthcoming, Transportation Research

Record

24. Xie, C., Lu, J., Parkany, E. (2003) “Work

Travel Mode Choice Modeling Using Data

Mining: Decision Trees And Neural

Networks,” Transportation Research

Record: Journal of the Transportation

Research Board, No. 1854.

http://trb.metapress.com/content/120399/



URBAN TRANSPORT

JOURNAL


* Research Scholar, Department of Civil Engineering, Indian Institute of Technology, ROORKEE 247667

INDIA, [email protected]

** M.Tech Student, Department of Civil Engineering, Indian Institute of Technology, ROORKEE 247667

INDIA, [email protected]

*** Associate Professor of Civil Engineering, Indian Institute of Technology, ROORKEE 247667 INDIA,

[email protected]

SELECTION OF ROUNDABOUT ENTRY CAPACITY MODEL FOR

INDIAN CONDITION

Abdullah Ahmad*, Srinath Mahesh** and Rajat Rastogi***

Abstract: Evaluating the capacity of roundabout is an important element in the planning and design of such

facilities. An empirical approach using regression analysis was used to develop a roundabout entry-capacity

model for Indian conditions. US model gave results close to that of field entry capacity model and hence US

model was calibrated for Indian condition.

Keywords: Roundabouts, critical gap, entry flow, circulating flow, entry capacity models

1.0 INTRODUCTION

A roundabout is an unsignalized

intersection with a circulating roadway and

connecting legs. It is a one-way circular

intersection without any traffic signal

equipment in which traffic flows around a

central island, clockwise for left-side driving

and anti-clockwise for right-side driving. It

operates with yield control at the entry points,

and gives priority to vehicles within the

roundabout. There is no specific way to

distinguish roundabouts from traffic circles or

rotaries. In the United States, these circular

intersections are classified into three

categories such as rotaries (traffic circles),

neighborhood traffic circles and roundabouts

(FHWA 2000). The fundamental difference is

in their design philosophies. Roundabouts

control and maintain low speeds for entering

and circulating traffic. This is achieved by

small diameters and low-speed entry

geometry. By contrast, rotary geometry

encourages high-speed merging and weaving,

made possible by larger diameters and large

high-speed entry radii. The geometric design

elements of roundabouts allow only slow

speeds therefore creating safer driving

conditions. A considerable number of

roundabouts have been installed in India.

Roundabouts have been used worldwide as an

efficient intersection control type to improve

safety and operational efficiency. Evaluation

of roundabout capacity is very important since

it is directly related to delay, level of service,

accident, operation cost, and environmental

issues.

For estimating entry capacity, the gap

acceptance approach and the empirical

approach are used now-a-days. The entry

capacity is considered as a function of

circulating flow as the circulating flow

decreases, the entry capacity increases due to

higher opportunities available for entering the

circulation area. The gap acceptance approach

estimates the entry capacity using critical gap

and follow-up time parameters. The empirical

approach estimates the entry capacity based on

the observed capacity of the existing

roundabouts which have been installed in early

days. Although the empirical models can best

reflect local traffic conditions, they cannot be

applied to other locations as such. Some of the

recent works in this area are as under

i. Çalişkanelli et al. (2009) applied regression

analysis method to compare the capacity

models. The data was collected at four

approaches of four multi-lane and seven

approaches of five single-lane traffic circles

in Izmir, Turkey. They found that the

method of critical gap acceptance gave

mailto:[email protected]


80

more accurate results than the other

models.

ii. Mazzella et al. (2011) considered a

geostatistical approach to establish the

relationship between entry capacity and

circulating flow. It emphasized that the

relationship between entry capacity and

circulating flow cannot be expressed by one

trend only but by two or three trends.

iii. Chandra and Rastogi (2012) proposed a

method to determine the entry capacity of a

roundabout in India. Data was collected at

four roundabouts in the suburban area of

Chandigarh city and analyzed using five

different methods of determining the entry

capacity. The proposed method gave the

capacity, quite comparable to the German

entry capacity model. Indian model (IRC -

65) gave the highest capacity amongst the

methods being considered i.e. UK, Swiss,

HCM and German model. However, Indian

model is based on the capacity of weaving

section which can accommodate the least

traffic. Among other four methods, UK

model gave the highest entry capacity and

the US model gave the lowest capacity.

iv. Dahl and Lee (2012) found that the

observed entry capacity was lower for the

roundabout with a higher truck percentage.

As truck percentage increased, the critical

gap and the follow-up time for the

roundabout increased, thus resulting in

lower entry capacities. The results showed

that the capacity decreased as truck

percentage increased, but the amount of

capacity reduction was less at higher

circulating flows.

In India, no comprehensive study is being

carried out to develop a model to estimate the

entry capacity of the roundabout. Few attempts

have been made to analyze the traffic flow and

estimate the entry capacity of the roundabout.

In this study, an effort is being made to

develop a regression model for estimating

roundabout entry capacity as a function of the

circulating traffic. The empirical approach,

using regression analysis, is used with this

purpose. The study is taken up as a part of

development of Indo-HCM. The objective of

this paper is to present and compare different

types of entry capacity models based on gap

acceptance and regression, and their

application to the settings of roundabouts in

the context of developing nations like India.

2.0 REVIEW OF ENTRY CAPACITY

MODELS

2.1.French Model(Guichet 1997)

The French formula for the estimation of

entry capacity (pcu/h), is based on the

regression analysis, and is given by equation

(1). This method considers the disturbing flow

in front of entry, follow up time and

roundabout geometries. The method is

somewhat complicated as the number of

variables and associated estimation formula

are more.

B dC *Q

3600C = A.e

(1)

Where, 0.8

e

f

W 3600A =

T 3.5

(2)

Tf = follow-up time (sec)

We = entry width (m)

CB = coefficient that is 3.525 for urban

areas and 3.625 for rural areas

Qd = disturbing flow in front of the entry

(pcu/h)

ud u a ci ti ce te

u c

QQ = Q .k . 1- + Q .k + Q .k

Q + Q

(3)

Qu = exiting flow (pcu/h)

Qc = Qci + Qce = circulating flow (pcu/h)

Qci = circulating flowon the far lane (pcu/h)

Qce = circulating flowon the near lane

(close to the entry) (pcu/h)

max

R L

R + W L for L <Lmax

Ka = (4)

0 for other cases

R= central island radius (m)


81

W = circulating roadway width (m)

L= splitter island width (m)

max

WL = 4.55 R +

2 (5)

ti

160k = Min or 1

W*(R + W)

(6)

2

te

(W -8) Rk = Min 1- or 1

W R + W

(7)

2.2.Jordan Model (Al-Masaeid and

Faddah 1997)

The model was developed based on

analysis using data from ten roundabouts. The

range of entry width and diameter of selected

roundabouts were 5-18 meter and 8-77 meter

respectively. The entry capacity model was

defined as a function of circulating traffic

flow, circulating width, entry width, diameter

of the central island, and distance between the

entry and its near exit. It is given by equation

(8).

c5.602 Q

0.312 0.219 0.071 EW + 0.019 RW 10000eQ =168.2 D S e e

(8)

Where,

Qe= entry capacity (pcu/hr)

Qc = circulating traffic flow (pcu/hr)

D = central island diameter (m)

S = distance between the entry and near-

side exit (m)

EW = entry width (m)

RW = circulating roadway width (m)

2.3.German Model (Brilon and Wu

2006)

The model for the estimation of roundabout

entry capacity was based on an idea from

Tanner (1967) as cited by Mauro and Branco

(2010). This is given by equation (9). The

method considers circulating flow, geometry

of the roundabouts and traffic flow micro

characteristics like gap, follow-up time and

headway.

cn

c e c fc

c f

.Q / 3600 n Q TC = 3600 1- * *exp - * T - -

n T 3600 2

(9)

Where:

C = entry capacity (pcu/h)

Qc = circulating flow in front of the entry

(pcu/h)

nc = number of circular lanes

ne = number of lanes in the subject entry

Tc = critical gap (sec)

Tf = follow-up time (sec)

∆ = minimum headway between the

vehicles circulating in the circle

2.4.US Model (HCM 2010)

This is an exponential model of entry

capacity for roundabouts. It is a combination

of simple, lane-based regression (exponential)

and gap-acceptance model. The roundabout

capacity model for an entry lane is expressed

as given by equation (10).

cB*VeC = A*e (10)

Where,

f

3600A =

t (11)

c ft 0.5* tB =

3600

(12)

Vc = conflicting flow rate in pcu/h

tf = follow-up time (s),

tc = critical gap (s)

3.0 DATA COLLECTION AND

EXTRACTION

For any traffic study, data collection is

extremely important and it is to be carried out

very carefully. The accuracy and care with

which the data collection is being carried out

in turn greatly affects the results. Therefore,

video recording technique was used to collect

the data at a roundabout. The use of a video

camera allows the collection of data with

minimum number of personnel and the video

tapes can be viewed several times to obtain the


82

multiple desired information. Study was

carried out for a selected roundabout in

Chandigarh, India. The video camera was

mounted on a stand and was placed on the roof

of a building located near the roundabout.

Recording was done for about 4 hours which

included morning peak period. The selected

roundabout was located in urban area and

there was no interference from pedestrians.

There were no parking and bus bays nearby

and roundabout was sufficiently away from

upstream and downstream signals. A snapshot

of selected roundabout is shown as Error!

Reference source not found.. The common

geometric features of the roundabout were:

a) Entry Width = 8.5 meter

b) Exit Width = 8.5 meter

c) Approach Width = 7.5 m

d) Departure Width = 7.5 m

e) Circulating roadway width = 7 m (2-

lanes)

f) Weaving length = 33 m

g) Central Island Diameter = 37 m

h) Central Island Perimeter = 115 m

i) Splitter Island = 3.5 m

Figure 1: Snapshot of selected roundabout in Chandigarh

In India, there is a mixed traffic condition

and during peak flows, lane discipline is not

followed. As entry driver tries to get space to

keep moving the gap acceptance behavior

becomes quite complex. Consequently, data

extraction for gap acceptance or rejection

process is a challenging task. The video was

played to extract the desired information. Data

on entry flow, circulating flow, accepted gap

and rejected gap by an entering vehicle were

recorded. Gap data were extracted with an

accuracy of 0.01 second. The categories of

vehicles found in urban area such as motorized

two-wheelers (2W), motorized three-wheelers

(3W), cars (CAR) and heavy vehicles (HV)

were considered for the analysis. The average

composition of traffic stream at entry flow and

circulating flow are shown in Table 1. The

entry traffic flow on an approach while having

a stable queue and the corresponding

circulating traffic flow were extracted for

period of queue dissipation ranging from 45


83

seconds to 4 minutes and were extrapolated to

the equivalent hourly flow.

Table 1: Composition of different types of

vehicles (%)

Flow

Conditions HV 3W Car TW Total

Entry flow 2 7 38 53 100

Circulating

flow 3 11 35 51 100

4.0 DEVELOPMENT OF ENTRY

CAPACITY MODEL

For estimating the entry capacity, the

circulating traffic flow and entry traffic flow

data were collected during periods of

continuous and stable queuing at the entry leg

of the roundabout. The entry capacity of the

roundabout will be the observed entry traffic

flow with a stable queue and a continuous

stream of vehicles with respect to the

circulating traffic flow (Al-Masaeid and

Faddah 1997). Circulating traffic flow and

entry capacity are expressed in passenger car

units (pcu) to account for two wheelers, three

wheelers and heavy vehicles. For conversion

into pcu, two wheelers are 0.75 pcu, three

wheelers are 1.0 pcu and heavy vehicles are

assessed as 2.8 pcu(IRC-65, 1976).

A scatterplot of entry capacity and

circulating traffic flow at roundabout is shown

in Figure 2Error! Reference source not

found..The relationship between entry

capacity and circulating traffic flow was

investigated and regression analysis was

carried out to determine the best fitted

equation by using entry capacity and

circulating traffic data. It is given in equation

(13). R square value for the exponential model

was higher than the R square value for linear

and other models. The relationship has been in

line with the reported literature (Al-Masaeid

and Faddah 1997; HCM 2000, 2010).

c0.0003*qeq 4752*e

(R

2 = 0.8907)

(13)

Where,

qe = entry flow (pcu/hr),

e = base of natural logarithm, and

qc = circulating traffic flow (pcu/hr)

5.0 ANALYSIS OF RESULTS

5.1 Critical Gap and Follow-up Time

Estimation

The critical gap is also extracted in this

case study for the comparison of the existing

entry capacity model as already discussed

above. Miller (1972) compared different

methods of critical gap estimation by using

simple gap acceptance model. The study found

that Maximum Likelihood technique and

Ashworth method gave acceptable results.

Maximum Likelihood technique was also

recommended by Troutbeck (1992). Brilon et

al. (1999) concluded that the Maximum

likelihood method and Hewitt’s method give

the best results. NingWu (2012) proposed a

method based on equilibrium of probabilities

for estimation of critical gap at unsignalised

intersection. Troutbeck (2014) compared the

Ning Wu method and Maximum Likelihood

Method. It was found that Maximum

Likelihood method was slightly better than

Ning Wu method. Among these methods, only

Maximum Likelihood Method proves to be the

most accurate and reliable. This method

requires data on both rejected gaps and

accepted gap by a vehicle. It utilizes the data

in pairs of highest rejected gap and next

accepted gap. Consequently, the critical gap

Figure 2: Entry flow versus circulating flow


84

values are estimated using the maximum

likelihood method for motorized two-

wheelers, motorized three-wheelers, cars and

heavy vehicles separately. These are given in

Error! Reference source not found.. Then

the critical gaps for the entire entry flow were

calculated as a volume-weighted average of

the critical gap for two- wheelers, motorized

three-wheelers, cars and heavy vehicles (Dahl

and Lee 2012).

Table 2: Critical gaps estimated by maximum

likelihood method

Critical Gaps (s) Weighted

Average

Critical

Gap (s) TW 3W CAR HV

1.59 1.92 2.12 2.51 1.83

Follow-up times represent the process by

which multiple vehicles that are queued at an

approach can enter the roundabout. In this

study, the extraction of follow up time was

very difficult. Therefore, the follow-up time

was taken as 0.6 times of the critical gap as

reported in literature (Brilon 1988; Hagring et

al. 2003; Tian et al. 2000). The extracted

critical gap and follow-up time for Indian

condition were used in French, German and

HCM (2010) entry capacity model.

5.2 Comparison of Capacity Models

The field entry capacity model was

compared with the already presented entry

capacity models. Error! Reference source

not found. shows the comparison between the

field entry capacity model and the other

models. Compared with French, Jordan,

German and the US entry capacity models for

roundabouts, the developed field entry

capacity model provided comparable estimates

with German capacity model only for 900 to

1400 pcu/h of circulating traffic. The field

entry capacity model gave the highest capacity

amongst all the methods i.e. French, Jordan,

German and the US model for circulating

traffic higher than 1100 pcu/h. The US model

gave higher capacity than French and Jordan

model for larger than 1100 pcu/h of circulating

traffic. Jordan model gave the lowest entry

capacity amongst all the methods. For

circulating volume less than 1100 pcu/h,

German and French model predicted higher

capacity than the other existing methods. The

trend followed by field data model in India

was comparable to that by the US model and

Jordan model. Field data model was higher

than the other two models. It signifies that

under mixed traffic conditions, the entry

capacity is higher than that under uniform

traffic condition.

Figure 3: Comparison of entry capacity models

Regression analysis has been done for

finding out the best capacity model among the

existing entry capacity models for Indian

condition. Error! Reference source not

found. to Error! Reference source not

found. show the field entry capacity model

versus existing entry capacity models to find

the best toning of the existing entry capacity

models with the field model. Relations

between existing entry capacity models and

field entry capacity model have been

developed. The linear relationships have been

taken into account rather than exponential or

other models for matching the style of the

existing entry capacity models with the field

model. The matching of French, Jordan and

German models with field entry capacity

model gave low R2 value as compared to the

US model. The best matching with the field

entry capacity model came out with the US

model for which R2 value was 0.9836. Based

on the R2 value, the US model complemented

with the field entry capacity model and can be


85

calibrated easily for Indian condition as given

below:

US Model 0.6764*FieldModel (14)

Field Model 1.478*USModel (15)

For practical application, the US model for

estimating entry capacity is simplified as

given below in equation (16) for Indian

condition.

cB*VeC =1.478*A*e (16)

Where,

f

3600A =

t (17)

c ft 0.5* tB =

3600

(18)

Vc = conflicting flow rate in pcu/h

tf = follow-up time for Indian traffic

condition (s),

tc = critical gap for Indian traffic condition

(s)

Figure 7: French model v/s Field model Figure 6: Jordan model v/s Field model

Figure 8: German model v/s Field model Figure 9: The US model v/s Field model


86

6.0 CONCLUSIONS

The following conclusions are drawn from

the study:

i. An empirical approach using regression

analysis was used to develop a roundabout

entry-capacity model for Indian

conditions. The entry capacity of an

approach of a roundabout is dependent on

the circulating flow in front of that

approach. As the circulating traffic

increases, the entry capacity decreases and

their relationship is found to be of negative

exponential nature.

ii. The circulating flow versus field entry

capacity charts were prepared for five

models. The field entry capacity model

gave the highest capacity amongst all the

methods i.e. French, Jordan, German and

the US model for circulating traffic higher

than 1100 pcu/h. Jordan model gave the

lowest entry capacity amongst all the

methods.

iii. By using regression analysis, the field

entry capacity model was compared with

other entry capacity models and it was

found that only US model gave best toning

to that of field entry capacity model for

Indian condition.

iv. The US model for estimating the entry

capacity is simple as compared to other

models since the number of variables and

associated estimation formula are less.

Henceforth, an adjustment factor was

applied to calibrate US entry capacity

model for Indian condition.

v. The critical gap and follow-up time values

recommended in HCM (2010) are not

applicable to Indian conditions where

smaller sized vehicles accept a much

lower gap and are able to force their entry

into the circulating roadway of the

roundabout.

The present study can be extended by

considering the following aspects:

i. To develop capacity charts for roundabouts having different geometric

condition like entry width, circulating

roadway width and diameter of central island.

ii. To estimate the dynamic passenger car

unit (pcu) values for all types of vehicles

which can be applied in the case of roundabouts.

REFERENCES 1. Al-Masaeid, H., and Faddah, M. (1997).

“Capacity of roundabouts in Jordan.”

Transportation Research Record: Journal

of the Transportation Research Board, No.

1572, 76–85.

2. Brilon, W. (1988). “Recent developments

in calculation methods for unsignalized

intersections in West Germany.”

Intersections without Traffic Signals,

Springer Berlin Heidelberg, 111–153.

3. Brilon, W., Koenig, R., and Troutbeck, R.

J. (1999). “Useful estimation procedures

for critical gaps.” Transportation

Research Part A: Policy and Practice,

33(3-4), 161–186.

4. Brilon, W., and Wu, N. (2006). “Merkblatt

für die Anlage von Kreisverke- hren

[Guideline for the design of

roundabouts].” FGSV Verlag Gmbh,

Cologne.

5. Çalişkanelli, P., Özuysal, M., Tanyel, S.,

and Yayla, N. (2009). “Comparison of

different capacity models for traffic

circles.” Transport, 24(4), 257–264.

6. Chandra, S., and Rastogi, R. (2012).

“Mixed traffic flow analysis on

roundabouts.” Journal of the Indian

Roads Congress, 73(1), 69–77.

7. Dahl, J., and Lee, C. (2012). “Empirical

estimation of capacity for roundabouts

using adjusted gap-acceptance parameters

for trucks.” Transportation Research


Research Board, No. 2312, 34–45.

8. FHWA. (2000). “Roundabouts: An

informational guide.” Federal Highway

Administration, Washington D.C.

9. Guichet, B. (1997). “Roundabouts in

France: Development, safety, design, and

capacity.” 3rd International Symposium


87

on Intersections Without Traffic Signals,

Portland, Oregon, USA, July 21-23, 100–

105.

10. Hagring, O., Rouphail, N. M., and

Sørensen, H. A. (2003). “Comparison of

capacity models for two-lane

roundabouts.” Transportation Research



11. HCM. (2000). “Highway capacity manual

2000.” Transportation Research Board,

National Research Council.

12. HCM. (2010). “Highway capacity manual

2010.” Transportation Research Board,

National Research Council.

13. IRC-65. (1976). “Recommendation

practice for traffic rotaries.” Indian Roads

Congress, New Delhi, India.

14. Mauro, R., and Branco, F. (2010).

“Comparative analysis of compact

multilane roundabouts and turbo-

roundabouts.” Journal of Transportation

Engineering, 136(4), 316–322.

15. Mazzella, A., Piras, C., and Pinna, F.

(2011). “Use of Kriging technique to study

roundabout performance.” Transportation

Research Record: Journal of the

Transportation Research Board, No. 2241,

78–86.

16. Miller, A. (1972). “Nine estimators of gap-

acceptance parameters.” 5th International

Symposium on the Theory of Traffic Flow

and Transportation, Newell, G. F. (ed),

American Elsevier Publ. Co, Inc., New

York.

17. Tanner, J. (1967). “The capacity of an

uncontrolled intersection.” Biometrika,

54(3), 657–658.

18. Tian, Z., Troutbeck, R., and Kyte, M.

(2000). “A further investigation on critical

gap and follow-up time.” Transportation

Research Circular E-C018: 4th

International Symposium on Highway

Capacity, Maui, Hawaii, June 27–July 1,

397–408.

19. Troutbeck, R. (1992). “Estimating the

critical acceptance gap from traffic

movements.” Physical Infrastructure

Centre Research Report 92-5, Queensland

University of Technology, Brisbane,

Australia.

20. Troutbeck, R. (2014). “Estimating the

mean critical gap.” Transportation

Research Board 93rd Annual Meeting,

Washington, D.C., January 12-16.

21. Wu, N. (2012). “Equilibrium of

probabilities for estimating distribution

function of critical gaps at unsignalized

intersections.” Transportation Research



URBAN TRANSPORT

JOURNAL


* Post Graduate Student, Civil Engineering Department, National Institute of Technology (NIT) Hamirpur,

Email: [email protected]

** Assistant Professor, Civil Engineering Department, NIT,Hamirpur, Himachal Pradesh, Email:

[email protected]

** Assistant Professor, Civil Engineering Department, College of Technology, G. B. Pant University of

Agriculture & Technology, Pantnagar (U. S. Nagar), Uttarakhand – 263145. Email:

[email protected]

FRAMEWORK FOR DEVELOPMENT OF ADVANCED TRAVELER

INFORMATION SYSTEM: A CASE STUDY FOR CHANDIGARH CITY

Bhupendra Singh*, Ankit Gupta**, Sanjeev Suman***

Abstract: Intelligent Transportation System (ITS) is an area of wide research in developed countries and a

lots of research work has been done in this area in the last two decades. There are many sub branches of ITS out

of which one of the most widely used worldwide is Advanced Traveler Information System (ATIS), which can

provide the information regarding the basic facilities to a traveler in a city.

Geographical information system (GIS) is a powerful tool for storage, graphical representation and analysis

of information of large data which makes it very useful for the development of ATIS. In this paper, a

comprehensive framework comprising of system architecture, development methodology, and salient features of

a GIS based ATIS for Chandigarh City, India has been discussed. The suggested system is able to provide the

information about the basic facilities of the city and help the users in planning and decision making about their

trips by providing shortest routes, nearest facilities and bus routes. This system can be stationed at public places

such as in KIOSK and used in personal computers at homes and offices.

Keywords: Advanced Traveler Information System, Intelligent Transportation System, geographical information system,

ArcGIS.

1.0 INTRODUCTION

Last two decades have seen a lot of

development in the field of transportation

infrastructure even then various traffic

problems are increasing day by day. This is

mainly due to the increase in number of

vehicles. Almost every country of the world

whether developing or developed, facing

problems in the management of transportation

facilities (Singh and Gupta, 2013). To solve

these problems the focus of countries is

shifting from the infrastructure development to

the optimum and best use of the already

constructed facilities and in this direction ITS

proves to be very useful. Intelligent

Transportation system is being utilized all over

the world to manage and solve different traffic

and transportation problems.

ITS is an integrated system that implements

a broad range of communication, control,

vehicle sensing and electronics technologies to

help in monitoring and managing traffic flow,

reducing congestion, providing optimum

routes to travelers, enhancing productivity of

the system, and saving lives, time and money.

ITS aims to improve the safety and efficiency

of the transportation system. ITS is a very big

area of study in itself containing lots of

subsidiary branches based on the use of them

in different traffic management fields, out of

which most important and widely used all over

the world to solve the traffic and transportation

problem are as follows:

Advanced Traveler Information

System (ATIS)

Advanced Traffic Management

System (ATMS)

Advanced Public Transportation System

(APTS), and

Emergency Management System (EMC).

Advanced Traveler Information System

(ATIS) implements a wide range of

technologies, such as internet web sites,

Framework For Development Of Advanced Traveler Information System: A Case Study For

Chandigarh City

90

telephones, cellular phones, television, radio,

etc. to assist travelers and drivers in making

informed decisions regarding trip departures,

optimum routes, and available modes of travel.

ATIS provides the drivers both en route and

pre-trip information which are advantageous in

many ways. Pre-trip information availability

enhances the self-belief of the drivers to use

freeways and allows commuters to make

better-informed transit choices (Campbell et

al., 2003). En route information and guidance

saves travel time, helps a traveler avoid

congestion, can improve traffic network

performance, and is more efficient than paper

maps or written instructions. In 1999 a survey

was conducted among the people who were

using the Advanced Regional Traffic

Interactive Management and Information

System (ARTIMIS) telephone traveler

information service in Cincinnati, Ohio. All of

them rated the service as beneficial service.

More than 99% of people surveyed in that city

said that they were benefited by avoiding

traffic problems, saving time, reducing

frustration, and arriving at destinations on time

and 81% said that they had recommended the

service to someone else. In this paper we have

discussed the methodology to develop an

Advanced Traveler Information System for the

Chandigarh city.


Advanced Traveler Information System is

being developed and used all over the world.

Most of the studies are based in the developed

countries and some are also based in the

developing countries also. Different platforms

and approaches have been used by the

different researchers. Some of the literature

work has been reviewed here:

Peng (1997) presented a method for

designing a geographic information systems

(GIS)-based automatic transit traveler

information system (ATTlS). The idea behind

the study was to provide the users optimal trip

option with least travel time between the

traveler's origin and destination, including

walking, waiting, transfer, and in-vehicle time

based on their origins, destinations, and bus

schedules and/or real-time information of bus

locations. To achieve the purpose of providing

the optimum route the methodology which was

adopted is to consider only those bus stop

points which are active (have service) at the

time of travel as all the bus stop points don’t

have the service all 24*7 and considering only

active bus stop point results in optimum route.

Wu et al. (2003) gave an ATIS based on the

Web service and wireless communication

technologies and in order to make the data

more reliable and useful for the commuters,

the methods of lost data reconstruction and

travel time prediction were also proposed and

examined in the study. An interpolation

method was used for the lost data construction

based on the periodical behavior of the traffic.

Travel time prediction is also done using

historical data based on the same observation

that the traffic possesses deterministic

behavior. The delivery medium which are used

in the system are mobile phones and personal

computers.

Kumar et al. (2005) developed a GIS based

advanced traveller information system for the

Hyderabad city, India under ArcView GIS

environment. The Avenue programming

language was used in the source program for

the system development. For the process of

Path optimization ArcView Network Analyst

(AVNA) is used. GIS-enabled modules for the

shortest path, closest facility, and city bus

routes were included in the system. The

developed system provides information about

basic facilities in Hyderabad City, such as road

networks, hospitals, government and private

offices, stadiums, bus and railway stations, and

places of tourist interests. The shortest path

facility developed in the system gives the user

full freedom to choose the origin and

destination either by themselves or by given

list in the system.


Chandigarh City

91

Hasnat et al. (2006) developed a similar

system using web and wireless communication

technologies. The system works in two

different modules 1. Web Based Service,

which provides the service to the user both in

text and map format. 2. SMS Based Service,

which receives the queries from the users in

the pre-defined format and then provide the

user information. Like most of the systems it

uses Dijkstra algorithm to compute shortest

path. In the computation of the total travel

time each road/ edge is given a weight based

on some constraints as traffic jam etc. and on

the basis of this the travel time is calculated.

Data Validation System is also included the

system to produce the data if any data source

stops working to produce the information

based on the historical data collected.

Singh and kumar (2010) presented an

overview of a web based ATIS for developing

countries keeping in mind that local traffic,

roadway, signalization, demographic,

topological, and social conditions in

developing countries are quite different from

those in developed countries. For the

development of system a Web GIS-based

architecture was adopted and customized and

termed as a Specific Design of Logical

Architecture (SDLA). SDLA is the specific

adoption of generalized three tier logical

architecture which is based on selected SW,

database design, and relations between SW

components comprising their in-between

interaction in terms of data flow and

information generation. Three tier architecture

used in the system includes:

1. Presentation tier which works as frontend

of the system to be used by the users to

make the quarries and get the results.

2. Application tier is concerned with the

processing of the data according to the

user’s need and sending this processed

information to the presentation tier to give

user requested information.

3. Data tier to store the data about the

different features.

The proposed system uses highway police

and traffic regularity authorities as main data

acquisition medium whereas desktop

computers and information kiosks as the

medium for distribution of information. The

work of processing of information is done

mainly with the help of GIS.

Pal and Singh (2011) gave a systematic

overview of a GIS system that can be used for

structuring, storing and dissemination transit

information for transit networks for

Metropolitan Cities in India which was

capable of handling real-time information.

Three-tier client-server SW architecture was

adopted as a logical architecture for

developing the ATIS. These three tiers which

were used are:

1. Presentation tier which works as user

interface

2. Application Tier as data processing and

information generation rules, and

3. Data tier for the handling of data (storage

and management).

Whenever a query is made by the user for

geospatial analysis web interface passes on

request to web server and web server then

passes on the request to a Geo server through

server connector. The Geo Server processes

client requests handed to it by the Web Server,

it accesses the spatial data, performs geo-

spatial analysis and renders web-ready map as

vector or raster image. The Data Tier is mainly

concerned with storage and management of

spatial information.

Zhang et al. (2011).in their study developed

and tested a generic multimodal transport

network model for ATIS applications. First, a

multimodal transport networks was modelled

from an abstract point of view and networks

were categorized into private and public

modes then a generic method was used to


Chandigarh City

92

construct a multimodal transport network

representation by using transfer links which

was inspired by the super-network technique.

For the computation of shortest path Dijkstra

algorithm is deployed. To check the

functionality of the developed model and

algorithm was tested based on a case study in

the Eindhoven region. The biggest problem is

that in the time taken for the route

determination by integration of different

different modes.

Mouskos and Greenfeld (1999) developed a

GIS-based MATIS which provided travelers

with access to information concerning route

planning by different modes i.e. private

automobile, mass transit, and ride sharing. The

System was developed under the ARC/INFO

GIS environment and census and graphic data

are acquired from the topologically integrated

geographic encoding and referencing (TIGER)

files from Union County, NJ.

3.0 STUDY AREA

For the development of the ATIS we have

chosen Chandigarh as our study area.

Chandigarh is first planned city of India

covering an area of approximately 44.5 square

meter or 114 km² and approximate population

of 1 million. Chandigarh has well maintained

roads and parking spaces all over the city ease

local transport. Chandigarh is considered to

have one of the best managed traffic and

transportation facilities in India but the

scenario is changing due to the increasing

number of vehicles in the city, this increased

population of the vehicles is causing

congestion and pollution on the roads.

According to the Centre for Science and

Environment (CSE) survey (2013):

1. Chandigarh has 441284 vehicles per 1000

km of road length whereas Delhi has

243783 vehicles per 1000 km of road

length.

2. Chandigarh has 227 cars per 1000 people,

whereas Delhi has 117 cars per 1000

(2011).

So we can see Chandigarh has almost

double vehicle density with respect to

population and road length as compare to

National Capital Delhi. So the need of a well-

developed ATIS is quite evident in the

Chandigarh which will help the users to avoid

congestion and spending less time on the roads

by providing them both en route and pre-trip

informations.

4.0 ATIS DEVELOPMENT

4.1. Methodology

Geographic Information System (GIS)

platform is used to develop the ATIS. Using

GIS environment to develop the ATIS offers

many advantages such as it allows large data

to be effectively processed, stored, analyzed,

logically associated, and graphical displayed.

ArcGIS 10.0 software developed by

Environmental Systems Research Institute

(ESRI) is used in the development of the

system. ArcGIS is a very powerful software

which provides geographic and spatial analysis

and can be used in different ways. In the study

the software will be used for the preparation of

different layers of the facilities of the

Chandīgarh city and their database. ArcGIS

10.0 also provides the facility to add

customized features through the programming

in Visual Basic for Applications (VBA). So

the features for giving the information about

the different tourist places of the city and bus

routes between a selected source and

destination will be added. So the proposed

system will be able to provide the following

informations

Shortest Routes among different places

Closest Facilities

Service Area of the facilities

Information about tourist destinations of

the city

Bus Routes between a source and

destination

4.2. Source Program:

Visual Basic for Applications (VBA) will

be used to write programs to add the


Chandigarh City

93

customized features in the software, VBA is

the implementation of the Microsoft’s

programming language Visual Basic 6. VBA

is widely used to develop user-defined

functions, automation and other low level

functionalities. Microsoft Access will be used

as the database for the customized programs.

4.3. Work Plan:

The flow chart of the work plan of the

proposed system is given in the following

figure 1:

Figure 1: Flow Chart of the Work Plan

4.4. Input Data:

The following data will be taken as Input

for the development of the system:

Map of the city having a Representative

Fraction of 1: 17250

Time Table of City bus Service.

Speed limits of the road.

Names of the roads.

Information of the other facilities.

Any type of the constraint on the roads

regarding direction and time.


Chandigarh City

94

4.5. Themes and Layers:

By the digitization of the basic map of the

Chandigarh city different layers of the

different facilities such as Roads, Hospitals,

Hotels, ATMs, Petrol Pump etc. have been

added in the system which provide the user

information regarding these facilities. The

Layers which have been added in the System

are:

Roads: approximately 5000 major and

minor roads have been added in the system

with the information such as length of the

roads, speed limit and name of the road (if

any).

Transport Facilities: In the transportation

facilities approximately 100 bus stops have

been included in the system with their

names. Railway station and Airport have

also included.

Educational Institutes: 170 government

and private school and colleges with their

names have been added in the systems.

Hospitals: 150 government and private

hospital with their names have been added

in the system.

Offices: 80 government and private offices

with their names have been added in the

systems.

Tourist Places: Main tourist attractions of

the chandigarh with their name, photos and

basic information are included in the

system.

Hotels: 180 hotels with their names have

been added in the systems.

ATM: 300 ATMs with the name of the

bank from which they belong names have

been included in the systems.

Petrol Pumps: 50 Petrol Pumps with their

names have been added in the systems.

Sectors: A separate layers to show the

names of the sectors is included in the

system so that the unknown user can easily

findout the sectors in the city.

It has been tried to include as much

facilities as possible to include in the system to

make it more user friendly. The digitized map

of Chandigarh city is given below in Figure 2:

Figure 2: Digitized Map of Chandigarh City

5.0 OUTPUT OF THE SYSTEM

The proposed system will be able to

provide the following functionalities to the

users:

1. Shortest routes: The proposed System

gives the user freedom to choose the point

between which he wants to know the

shortest route. User can add more than two

points also giving them numbers and the

system will give the shortest routes

between these points according to the

numbers from point 1 to 2 to 3 etc. User

can also shuffle these points to change their

number. The system provides the shortest

routes based on two parameters i.e. distance

and time. The direction window of the

system provides the direction instructions

to reach the destination. Figure 3 shows the

shortest routes between three selected

places:


Chandigarh City

95

Figure 3: Shortest Route between three

2. Closest facility: There are many facilities

included in the system such as hospitals,

offices, hotels, educational institute etc. So

if the user want to know any closest facility

then he can choose the desired facility from

the list and mark his position on the map,

based on these two parameters the system

will provide the closest facilities available

near the location of the user. The closest

facility is given based on the distance or

time parameter set in the system. These

parameters can be increased or decreased

based on the users need. Below Figure 4

shows the closest facility function of the

system:

Figure 4: Closest Facility

3. Service area of the facilities: In the

proposed system with the help of the

service area option the user can also find

out the service area of the facilities. The

desired service area can be modified with

respect to time and distance parameters

based on the users need. The shaded area in

the Figure 5 shows the service area of the

selected facility:

Figure 5: Service Area

4. Information about tourist destinations of

the city: This is the customized option

which is added in the proposed system. In

this option user will be able to choose

desired tourist place from a given list and

then click give details button, clicking the

button will show the picture and basic

information about the place. Figure 6

shows the information form of the tourist

places in the city:

Figure 6: Information Form of the Tourist

Places

5. City bus routes: This is another

customized feature added in the system.


Chandigarh City

96

With the help of this feature the user can

choose his origin and destination from the

given lists of different bus stops in the city

and based on origin and destination the

system will give the desired city bus

number and its timing.

6.0 CONCLUSIONS

The proposed GIS based ATIS can give

user some useful informations such as closest

facility his location, shortest route between his

origin and destination, City bus routes etc.

these informations can be very useful for the

planning of the trip. If the user’s trip will be

well planned then he has to spend less time on

the roads which will save the fuel cost and also

checks the adverse pollution effects due to

vehicles. The System is useful to the users

who are relatively new to the city, they can use

the system to get the information about of the

basic facilities of the city. Chandigarh is being

the capital city of two states is a popular

tourist destination in north India region, so the

developed system can prove to be very useful.

The System has a simple user interface.

Many day to day needed basic facilities such

as petrol pumps, ATMs, Hospitals etc. are

added in the system to give user information

about these facilities. The developed package

can be used at different places areas in the

KIOSKS to give information to the travelers,

such as:

Bus Stands;

Railway Stations;

Airports;

Tourist Information Centers;

In Personal Computers;

In cars.

The proposed system is the basic system

giving information about the basic facilities of

the city. The system can further be developed

to give various real time informations such as

the current congestion situation on a certain

road, current location of the desired city bus

etc. The system can be modified to provide

internet facility to the user i.e. it can be

modified to the internet GIS based ATIS. Also

as the system provides only the information

about city buses, different modes of

transportation can also be added to the system

to make it more user friendly

REFERENCES

1. Campbell, J. L., Carney, C., and Kantowitz,

B. H., (2003) “Human Factors Design

Guidelines for Advanced Traveler

Information Systems (ATIS) and

Commercial Vehicle Operation (CVO),”

Federal Highway Admin., McLean, VA,

Rep. FHWA-RD-98-057-2.

2. Hasnat, M. A., Haque, M. M., and Khan

M., (2006), “GIS Based Real Time Traveler

Information System: An Efficient Approach

to Minimize Travel Time Using Available

Media”, available from

http://www.bracu.ac.bd/directories/www/?c

ode=56014.

3. Kumar, P., Singh, V., and Reddy, D.,

(1999), “Advanced Traveler Information

System for Hyderabad City”. IEEE

Transactions on Intelligent Transportation

Systems, VOL. 6, NO. 1, pp. 26-37.

4. Mouskos, K. and Greenfeld, J. (1999) “A

GIS based multi modal advanced traveler

information system (MATIS)”, J. Comp.

Aided Civil Infrastructure Eng., vol. 14, no.

4, pp. 267–279.

5. Pal, S. and Singh, V., (2011), “GIS Based

Transit Information System for

Metropolitan Cities in India”, in the

proceedings of Geospatial World Forum,

18 – 21 January 2011, Paper Reference

No.: PN-250

6. Peng, Z. R., (1997), “A Methodology for

Design of a GIS-Based Automatic Transit

Traveler Information System”, Comput.,

Environ. and Urban Systems. Vol. 21. No.

5, pp. 359-372.

7. Roychowdhury, A., Bansal R., and

Chattopadhyaya, V., (2013) “Air Quality

http://www.bracu.ac.bd/directories/www/?code=56014

http://www.bracu.ac.bd/directories/www/?code=56014


Chandigarh City

97

and Urban Mobility Challenges”, Centre

for Science and Environment (CSE),

Proceedings of Workshop on Clean Air and

Sustainable Mobility, Chandigarh, May 24,

2013.

8. Singh, B. and Gupta, A. (2013),

“Advancements in Intelligent

Transportation Systems: A Review”, Proc.,

of International Conference on Advance

Trends in Engineering and Technology,

ICATET-2013, Jaipur, India, pp. 375-383.

9. Singh, V. and Kumar, P., (2011), “Web-

Based Advanced Traveler Information

System for Developing Countries”, Journal

of Transportation Engineering, ASCE, Vol.

136, No. 9, pp. 836-845.

10. Wu, C. H., Su, D. C., Chang, J., Wei C. C.,

Ho, J. M., Lin, K. J., and Lee, D.T., (2003),

“An Advanced Traveler Information System

With Emerging Network Technologies”, in

the proceedings of 6th Asia-Pacific Conf.

Intelligent Transportation Systems Forum.

11. Zhang, J., Liaoa, F., Arentzea, T.,

Timmermansaa H., (2011), “A Multimodal

Transport Network Model for Advanced

Traveler Information Systems”, Procedia

Computer Science 5 (2011), pp. 912–919.

URBAN TRANSPORT

JOURNAL


* Urban Transport Planner, Institute of Urban Transport (India), [email protected]

** Urban Transport Expert, Institute of Urban Transport (India), [email protected]

INSTITUTIONAL AND FINANCIAL STRENGTHEINING OF

INTERMEDIATE PUBLIC TRANSPORT SERVICES IN INDIAN

CITIES

Anindita Ghosh* and Kanika Kalra**

Abstract: Within the urban transport framework, intermediate public transport (IPT) like 3wheelers, auto

rickshaws, tempos and Tata Magic caters the daily urban trips in Indian cities. In absence of an organized city

bus service they provide an alternative mode of travel and where Public Transport is available, they act as a

feeder to the system. However, due to the unorganized nature of this sector, it faces many challenges. The recent

recommendations of the working group on urban transport, both for the 12th Five Year Plan and the NTDPC

stresses the need to improve the IPT Services as these vehicles have a potential of providing clean mobility and

low emissions solutions. This paper focuses on the major issues faced by this sector and the solutions to

organize and regularize the system in Indian cities.

There are several challenges faced by this sector. However, all the issues or the challenges identified may be

addressed by organizing the IPT under the umbrella of an existing SPV or setting up a new SPV, in case an

existing SPV is not available. This would not only organize the existing/ new IPT services but at the same time

would not impose heavy financial burden on the government.

Keywords: Intermediate Public Transport, Mobility, Intelligent Transport System, socio-economic stability, Special

Purpose Vehicle

1.0 INTRODUCTION

India is experiencing rapid urbanization

and motorization. While the urban population

is growing at the rate of 3.16 % per year,

motor vehicles are growing at a rate of 9%.

(Sharma, Jain, and Singh,2011). Today, buses

constitute less than 1% of the total registered

vehicles in cities (Road Transport Yearbook,

2011-12). In fact, very few Indian cities have

organized, regularized and regulated public

transport system in the absence of an

organized city bus service, the gap is being

filled by intermediate public transport (IPT)

modes like 3-wheelers auto-rickshaws,

Tempos and Tata magic etc which provide

public transport services (India Transport

Report- Moving India to 2032,2014).

Recent policy initiatives by the Central

government, such as the 2005 Jawaharlal

Nehru National Urban Renewal Mission

(JnNURM) and the National Urban Transport

Policy (NUTP), 2006 aim to provide a vision

and framework for promoting sustainable

urban transport in India. However, neither the

programme and nor the policy focuses on the

important role of IPT and its need for

improvement and up gradation. Though the

recent recommendations of the working group

on urban transport both for the 12th Five Year

Plan and the NTDPC stresses the need to

improve the IPT Services as these vehicles

have a potential of providing clean mobility

and low emissions solutions. Therefore the

need arises to improve and upgrade the IPT

vehicles and services, recognizing the

important role they play in the Indian Cities.

This paper focuses on the major issues faced

by this sector and the solutions to organize and

regularize the system in Indian cities.


The concept of Intermediate public

transport (IPT) differs in the context of

developed and developing countries. In

developed countries, IPT is often used as a

demand responsive system such as shared-ride

taxis and dial-a-ride services. In case of

Institutional And Financial Strengthening Of Intermediate Public Transport Services In Indian Cities

100

developing countries, lower standard of living,

high population density and easy availability

of cheap labour force have together provided a

variety of transport modes fulfilling the gap

between public transport and private vehicles.

It provides with several benefits like- mobility

and connectivity, market responsive services

and low cost travel option. Depending on a

city’s size and transport characteristics, IPT

modes may fall under two broad categories

:(1) contract carriage services, which are

flexible demand-based services where the

passenger determines the destination, and (2)

informal public transport (bus like) services,

characterized by fixed-route services with

intermediate stops for boarding and alighting.

Both kinds of services exist in India.

This sector faces tremendous challenges in

Indian cities due to their un-regularized nature

of operations. Case studies from the

developing countries like minibus taxis of

South Africa, the IPT system of Indonesia,

Dolmus of Turkey, G-Auto of Ahmedabad etc,

were referred, to understand the initiatives

undertaken by these cities to improve the

services. All the case studies indicate that the

key ingredients for a sustainable IPT system

are:

Strong regulatory authority fixing the

routes, fares, laws

Provision of proper infrastructural facilities

like parking areas, stands, separate lanes

etc,

Provision of financial and social benefits

to drivers through various government

schemes and

Usage of modern technologies is some of

the ways to organize the system.

3.0 METHODOLOGY

The methodology for the study is broadly

divided into three stages, i.e., literature review,

field visit and recommendations. In the

literature review stage, the basic challenge

faced by the IPT sector in various cities and

the measures taken to address the challenges is

understood. The first stage was complemented

with a study of 19 Indian cities, for which field

visits were made to understand the ground

conditions. The second stage was marked by

discussion of the questionnaire with more than

30 city officials (RTO and Traffic Police) and

primary survey of more than 1,900 drivers/

auto unions, in 19 cities(selected based on

population size)across India (Refer Table 1).

The sample size considered for the primary

survey varies between 0.1%-1%, depending on

the total number of registered IPT vehicles in

the city. The analysis of cities was done based

on the three categories listed in the table

below. The discussions with the stakeholders

led to an understanding of the existing system

and identification of gaps and problems in the

sector. This is followed by the final stage of

the study, which provided suggestions and

recommendations for improving the system.

Table 1: List of Cities Selected for Study along

with Population Size

S.

No.

Population

size

Number

of cities

Name of the

city

1 5-10 lakh 5 Guwahati,

Chandigarh,

Jammu, Alwar,

Kochi

2 10-20 lakh 6 Bhopal, Indore,

Ghaziabad,

Jodhpur,

Ranchi,

Amritsar

3 20 lakh and

above

8 Lucknow,

Kanpur, Surat,

Ahmedabad,

Kolkata, Delhi,

Mumbai,

Bangalore

4.0 EXISTING SCENARIO

In Indian cities the key role played by IPT

are of two types


101

In smaller and medium size cities, the

dominant mode of public transport like

Alwar, Amritsar etc

In larger cities, it acts as a feeder to the

main mode of public transport like metro,

BRT, suburban rail etc. Example are Delhi,

Mumbai, Kolkata and Ahmedabad

4.1 Composition

The predominant type of IPT vehicles

across Indian cities is 3 seater auto rickshaws

(60%), followed by Tata Magic (4 wheeler IPT

vehicles with seating capacity of 8 passengers)

24%, as many cities like Delhi, Bhopal,

Indore, Alwar in order to comply with the

emission standards are switching to a higher

grade vehicles and Tempos/Vikram (3 wheeler

IPT with a seating capacity of 6-8 passengers)

constituting only 16%.

4.2 Policy/Acts and Rules

The existing National Transport Policy

2006 does not focus on the role of IPT in

Indian cities. The existing Central Motor

Vehicle Act 1988 and the State Motor Vehicle

Rules also do not focus on the role and the

institutions for the enforcement of various

duties and responsibilities related to motor

vehicles other than identifying the Regional

Transport Authority (RTA) as the registering

authority for all vehicles including IPT

vehicles. The act does not provide the method

for fixing the routes and fares, use of modern

technologies to improve performance of IPT

vehicles, mode of financing the vehicles,

improvement of socioeconomic conditions of

drivers, etc. Therefore, there is a need to revise

the existing policy, act and rules.

4.3 Permits

In India the regulatory authorities for IPT is

the RTA, which issues permits and licenses to

the drivers and the traffic police, which is

responsible for enforcement of rules and

regulations on roads. The documents required

for obtaining permits are more or less similar

for all cities - application form, residence

proof, driving license, fitness certificate, PUC

etc.

It has been also observed that the permit

system in India is of two types(1) open permit

system, with no cap on the number of permits

issued, like in case of Surat and (2) closed

permit system, where there is a cap on

permits, like Mumbai, Kolkata, Delhi etc.

However, the cap on the number of IPT

vehicles is ad hoc, which results in a large

number of unauthorized vehicles operating in

cities like Kolkata, Lucknow, Delhi etc.

4.4 Lack of ownership / institution

for the IPT vehicles

The Central Motor Vehicles Act does not

recognise the institution responsible for the

discharge of its function and responsibility

towards the IPT vehicles, other than the RTA

for issuing permits and traffic police for

looking after the rules and regulations on

roads. This sector is considered to be

unorganised and completely privately owned.

Therefore, no recognition is given by the

government in organising the system like

improving the fleet, financing the vehicles and

improving the working and social conditions

of the drivers.

4.5 Routes

The routes of the 3 seater auto rickshaws

(80%) are generally not fixed for operation by

the RTA, except in the case of Guwahati and

Kolkata. In case of vikrams /tempos (70%)

routes are mostly fixed by the unions and the

drivers themselves. Lack of proper route

rationalization often results in greater

competition between drivers, rash driving

practices and inappropriate distribution of

services in the city,

4.6 Fare Fixation

There are no fixed rules for fixation of

fares. In case of the 3seater auto rickshaws the

fares are fixed by the RTA on the basis of

Government notification. In case of cities

where tempos /vikrams/ Tata Magic operate,


102

70% of the cities do not have fixed fare

system, the fares are decided by the unions and

the drivers themselves. It can also be noted

that due to lack of standardized analytical

framework for fare determination,

implementation and revision, there is

overcharging by drivers and conflicts between

drivers, union’s, commuters and authorities.

4.7 Infrastructure Facilities

In most cities, adequate number of IPT

stands, interchange and parking facilities for

the vehicles are not provided. As a result, these

vehicles queue along roadside leading to

congestion, especially near the junctions. In

some cities like Jodhpur and Ahmedabad,

though stands have been notified by the Nagar

Nigam and Municipal Corporation, due to lack

of enforcement the stands and interchange

facilities are often encroached by the hawkers.

Other infrastructures like gas stations,

registered repair shops and rest rooms and

shelters for drivers have also not been

provided.

4.8 Vehicle Technology and Fuel Type

The city size does not have a bearing on the

type or characteristics of IPT vehicles. It has

been observed that 64% of the vehicles across

Indian cities are 4-stroke and 2 stroke accounts

for only 36% of the total IPT vehicles. 2 stroke

vehicle types are mostly found in category 1

and 2 size cities due to their lower capital and

maintenance cost. This also results in high

levels of pollution.

The predominant fuel type used by IPT

vehicles across India is CNG/ LPG (60%).In

the remaining 40% of the cities, mostly

belonging to category 1 and 2, a blend of

diesel and petrol vehicles are being operated.

This results in greater levels of pollution and

greenhouse gas emissions.

4.9 Use of ITS in Vehicles

IPT, unlike cabs and private vehicles, do

not use modern technologies like GPS, panic

button, etc. As a result, these vehicles are

usually concentrated in a place where the

probability of getting passengers is the highest.

Secondly, the traffic police often penalize the

drivers for not wearing uniforms, non-usage of

fare meter, violation of routes, lack of

documents etc without giving a proper challan

to drivers. Thirdly, IPT is not considered as a

safe mean of transport especially for the

females and elderly people as the vehicles

cannot be tracked and lastly the drivers often

charge illegally as there is no fixed meter

system in most cities.

4.10 Financing of IPT vehicles

Most of the drivers rent the vehicles from

their owners as they are financially weak and

the process of getting loans is not favourable

for them. In India, the nationalised banks lend

money at 12.5% to 15.5%, however the

applicant needs to submit many documents

like address proof, pan card, etc. In the

absence of easy loans from nationalised bank,

most (about 75%) of the drivers resort to the

private bank’s and money lenders for funding

or take the vehicle on rent for operations.

Although these banks have a higher rate of

interest (20 to 25%) their requirements in

terms of documents is much lesser, and offer a

faster procedure for sanction of loans.

4.11 Socio-Economic Condition of the

Drivers

From the survey, it was observed that about

2/3rd of the drivers working in this sector have

only completed primary education. Most of the

drivers surveyed (70%) did not own the

vehicle and were operating them on rent. The

average rent paid per day by a 3 seater auto

rickshaw driver is Rs 250 and Rs 650 for Tata

Magic/ Vikrams. Considering that the average

kilometres driven by the drivers in cities are

approximately 100 kilometres (based on on

primary survey), and the average ridership is

45passengers/day(based on primary survey),

the revenue earned by drivers per month is

given in the table below:


103

Table13: Revenue Earned/ Month- General

Services

Description Earning/

month after

deduction of

rent (Rs)

Average

earning/

month

(Rs)

General Services

Earning without

rent

16,000-

18,000

17,000

Earning with rent 10,000-

15,000

12,500

Shuttle Services/ Shared Services

Earning without

rent

18,500-

20,500

19,500

Earning with rent 10,000-

20,000

15,000

The above analysis indicates that nearly

25% of the earnings of a driver accounts for

payment of rent. Further, the average

maintenance cost of each 3 wheeler auto

rickshaw is Rs 1,700 per month and for

Vikram /Tata magic it is Rs 1,300 per month.

Other miscellaneous expenses incurred by a

driver are Rs 300/ month (based on primary

survey).Considering the expenses incurred by

a driver on daily maintenance, fuel and rent,

the average monthly saving of a driver is Rs

7,000 per month (Refer, Table 3)

Table 14: Total Savings/ Month

Description Average

Earning/

month (Rs)

Average

Maintenance

cost /

month(Rs)

Average of other

expenses( bribes,

membership fees

etc)/month (Rs)

Average

Fuel cost/

month (Rs)

Total

Savings /

month (Rs)

General Services

Earning without

rent

17,000 1,700 300 6,500 8,500

Earning with rent 12,500 1,700 300 6,500 4000

Shuttle Services/ Shared Services

Earning without

rent

19,500 1,300 300 7,800 10,100

Earning with rent 15,000 1,300 300 7,800 5,600

This has a significant impact on the

financial status and overall well being of the

drivers. In some cases the earning is even

lower than the minimum wages specified by

the various states Labour Departments. Thus it

leads to greater economic instability among

drivers and his family.

This sector being unorganised in India, the

drivers work individually. It is very rare that

they get any social benefits from the unions, or

from NGOs. They are also not provided help

from the government in the form of training,

insurance, medical facilities, pension,

education, etc, except for few social groups.

Also to earn their daily wages and to cover the

operating expenses the IPT drivers work for

more than twelve hours a day, resulting in

constant exposure to on road pollution and

poor health conditions. This also leads to

weakness, tiredness and thus unsafe driving

practices.

4.12 Users Perspective

From the survey, it has been observed that

more or less all age groups of people use the

IPT system, however, commuters in the age

group of 30-40 years or are predominant. The

average distance travelled by the passengers is

approximately 5.5 kms, with an average

spending of Rs 600 per month. Some of the

major issues faced by the users are, high fares

being charged by the operators as the fare

meters do not work, absence of dedicated auto


104

rickshaw stands and parking areas which,

often leads to chaos and congestion on roads,

overloading in case of shared services, safety

and security issues, especially for female and

elderly users and non-availability of auto

rickshaw services at night time.

A preference survey was done for drivers

and users regarding the improvement of the

system like providing of infrastructure like

stands, usage of modern technology for

improvement of security, fixation of fare

structure for IPT, etc. About 80% of the users

are of the opinion that such recommendations

should be included in order to upgrade the

system.

Thus, it can be said that though IPT is an

important mode in all cities across India, this

sector being unorganized and not recognized

by the government are facing some of the

biggest challenges in the times to come.

Therefore, in order to upgrade the service and

IPT vehicles there is a need for improvements

in the sector before it becomes too late and

their importance is lost.

5.0 RECOMMENDATIONS

Some of the recommendations that can be

followed by cities to solve the challenges are

given in below:

5.1 Policy and Regulatory

Framework

A. Policy

The role of IPT needs to be clearly defined

in the National Urban Transport Policy, 2006

(NUTP) as follows:

In cities where there is no proper network

of public transport like Bus, Metro, and

Rail, the IPT should act as a feeder to the

main modes.

In case of cities where public transport is

completely absent, the IPT can act as a

main mode till a suitable PT option is

developed for the city.

In cities whether the PT system has a

skeletal framework, the IPT should act as a

complementary mode in only those areas of

the cities where public transport supply is

not available and other areas it acts as the

feeder to main PT services.

B. Acts and Rules

There is a need for the review of the

Central and State Motor Vehicles Rules. The

central act should include the roles and

responsibilities of various institutions,

standard clauses for the State Motor Vehicles

Rules relating to issue of permits, penalties,

and time for processing, documents required.

The state rules should also indicate the

various kinds of fees to be paid during the

issue of permits. Regarding the issue of

permits, the state can put a cap on the numbers

of IPT vehicles plying within the city. Further,

route rationalization and fares revision also to

be included in the rules.

C. Institution for IPT

In order to solve most of the problems and

the challenges faced by this sector as stated

earlier the IPT services should be organized

under the umbrella of an existing Special

Purpose Vehicle (SPV) ora new SPV may be

set up, in case an existing SPV is not available.

This would not only organize the IPT services

under an umbrella organization but would also

not have any heavy financial burden on the

government.

The operations of the vehicles can then be

done on Public Private Partnership (PPP)

basis.To ensure a good quality of service the

SPV while selecting the operator can specify

the number of permits, routes, fares,

technology, emission standards, performance

standards, marketing, training for drivers, ITS

facilities etc for the system.

To manage the operations, the SPV will

consist of some staff taken on deputation from

the government, whereas others will have to be

recruited from the open market as the required

skills would not be available with any agency


105

of the government. Apart from these, other

members who can be a part of this SPV should

be the RTA, an engineer from Municipal

Corporation who is responsible for the

provision of infrastructure and representatives

from the traffic police.

D. Permits

In order to bring consistency in issue of

permits the steps that can be taken include

developing a single Motor Vehicle Rule for all

states, variation in the permit fees for IPT

based on the type of route (Refer Table 5)and

fixing the number of permits to be issued by

the city as per the Table 4 given below. This

would ensure that the supply of IPT vehicle

meets the demand and at the same time

controls over supply, which may result in

unhealthy competition and congestion on the

streets. The following are the assumptions for

calculating the number of permits under

various scenerios:

Seating capacity of standard bus is assumed

to be 40 passengers per bus (based on

inputs from bus manufacturers).

Assumption are depended on the kind of

existing public transport in the city.

23% of users use IPT as a feeder to main

mode. (Business plan for operations of

feeder services DMRC, (2009).

It is assumed that only 3 seater auto

rickshaws operate in megacities.

Table 4 Number of Permits to be Issued under Various Scenarios

SL.

No

Scenarios As % of the

seating capacity

of the bus

Tempos or Tata

Magic (8 seater) / lakh

population

Auto rickshaws (3

seater) / lakh

population

Medium and small size cities

1. Absence of Public

transport

100 % 200 approx 500 approx

2. Presence of a skeletal

form of Public

transport

25%-70% 50-140 approx 125- 350 approx

3. Presence of Public

transport (acting as

feeder services)

23% 50 approx 120 approx

Megacities more than 4 million population

1. Presence of Public

transport (acting as

feeder services)

23% N.A 140 approx

*Table above is just for calculating the ratio; however it has to be adjusted according to the city’s existing fees

structure.

E. Route Rationalization

Following are the three options that can be

used

In order to take care of the problem of route

rationalization 2 steps may be taken to

develop routes of shared IPT. First is the

modification of the existing routes in order

to delete maximum overlaps and

competition and second would be to

introduce new routes in areas where IPT is

presently absent. This would help in

bringing efficiency and reliability in the

system for commuters and reduce

competition leading to better earning

amongst drivers. This, along with strong

enforcement by the traffic police would

also help with proper implementation.

The second option can be route wise permit

fee variation for the IPT services. To end

the problem of lack of service coverage in

few areas of the city and also to incentivise


106

the drivers with low financial status, fees

may vary according to the demand of areas.

Higher the demand of a route, more could

be the permit fee and vice versa. Permit fee

can thus be fixed as per Table 5 given

below.

Table 5 Estimated Route Permit Fees

Minimum (Rs)

(Less profitable routes)

Average (Rs) Maximum (Rs)

(Profitable routes)

Earning /day by

drivers(30 working

days)

200 600 1000

Earning /month by

drivers

6,000 15,000 24,000

Earning / year by drivers 72,000 2,16,000 3,60,000

Earning over 5 years by

drivers

3,60,000 10,80,000 18,00,000

Ratio of permit fees 1 2.5 5

Price of the 5 years

permit

200 500 1,000

*For implementation if the amount is considered to be too less, then a factor of maybe 20% can be added to

the fees.

A third option that can be adopted is

clubbing the profitable routes with the

non profitable routes and developing a

cluster system in which operations for

certain routes are tendered to private

operators, so that every operator in its

own cluster has both the routes to

equalize the variation in earning.

All these methods are possible only if an

SPV is set up to regularize IPT operations.

5.2 Infrastructure facilities

Following are the suggestions for

improvement:

Halt and go facilities to be developed

along roadside and interchange facilities

should be created near the bus stands and

metro stations depending on the demand

of passengers and land use locations

surrounding it.

Parking areas should be identified for the

drivers to safely park their vehicle at

night on payment basis. The same area

should also be provided with common

repair and maintenance facilities.

Creation of new auto stands with various

amenities like rest shelters, drinking

water and toilet facilities to improve the

working conditions of drivers.

Setting up of gas stations seems

essential.

There should be strict enforcement for

regulating the stopping and halting of

IPT vehicles near intersections. A

minimum distance of 250 meters from

the intersections/ junctions should

observe for restricting the stopping and

halting of IPT.

5.3 Technological Up gradations

In order to solve problem of outdated

technology and to meet emission standards

the following suggestions are made:

In order to replace the old polluting

vehicles on the roads, the government

plying needs to provide financial

incentives such as sales tax exemption

and interest subsidy on loans, for

retrofitting latest technologies. Few cities

like Delhi, Bangalore, Kolkata, Chennai

and Hyderabad have already taken the


107

lead in this. As a fiscal incentive for

CNG/LPG conversion city government

provided a subsidy of around Rs 2,000 to

3 seater auto rickshaw owners.

With the financial incentives from the

government the drivers and

manufacturers must be encouraged to

upgrade their IPT vehicles to 4 stroke

and to BS IV standards, since the new

BS IV standards are to be launched in the

year 2015.

Regulatory measures should be put in

place for thecreation of a single nodal

agency specifying the standards and

norms, keeping in mind the latest

technologies of vehicles, adoption of

separate emission standards for HC and

NOx emissions and defining CO2

emission standard.To implement the

emissions legislation set up by the

Ministry of Road Transport and Highway

(MORTH), state governments should

restrict the age of IPT vehicles to a

maximum of 8 years, so that it runs in

good condition.

Other measures include setting up of

more CNG/LPG stations and more

research into alternative fuel and vehicle

technology in order to lower the cost of

the vehicle.

Use of modern ITS technology could

significantly upgrade the service and solve

many issues like unequal dispatch of

vehicles, security to passengers,

enforcement by traffic police and

overcharging by drivers from the

commuters. Some of the measures are:

For more efficiency, the various

components of ITS can be implemented

in two phases. In phase 1 the panic

button and GPS can be installed on the

vehicles along with a Traffic

Management Centre to monitor the

movement and dispatch of vehicles. The

auto rickshaws may also be installed

with “hired/vacant” panel (status panel),

and E-challan may be introduced. The

phase 2 would consist of the

implementation of the Passenger

Information System (PIS), Security

Camera and smart-card reader.

For installation of the ITS devises on

existing vehicles, subsidy may be

provided by the state/ city government to

the owners of these vehicles to partially

meet the cost of GPS/GPRS. It could be

similar to the case of Delhi NCR where it

has already been made mandatory for the

drivers to have a GPS in order to get

their fitness certificate. All new vehicles

may be preinstalled with ITS devices and

the government may define the

specifications for them as has been done

for urban buses in India.

The control centre can be set up on a PPP

basis where the private party can recover

its cost on a monthly basis through the

extra transaction cost that can be charged

to the passengers along with the fares. A

similar system has already been

implemented by G-Autos in Surat, where

Rs 20 is being charged as the transaction

fee. Alternatively, the existing control

room for the public transport system can

be integrated with the IPT vehicles.

5.4 Economic/social stability for

drivers

In order to bring economic stability the

following recommendations are made:

The maintenance cost can be lowered by

providing the drivers a shared repair

workshop along with a proper training to

do the basic repairs, so that every time

for minor repair work the drivers do not

have to go the private workshops.

In order to increase the revenue for the

drivers, other options like

advertisements, renting of the vehicles

for rallies, to schools, to the tourism

department etc can be explored in

various cities.


108

In order to solve the problem of

financing the IPT vehicles for the

drivers, the most appropriate option

would be institutionalizing the services

under the umbrella of an SPV and

tendering the operations of the vehicles

on PPP. The drivers can form a

consortium and bid for the services. The

finances would be much more easily

available to the consortium as compared

to individual drivers, as they will be

known by the SPV/ government.

Besides, there would not be any risk of a

sudden shortage of funds or close down

of companies, therefore the system

would continue to work regularly and

provide economic stability to drivers.

As given earlier, the average savings of

an IPT driver are less than Rs 7,000 per

month, i.e. about Rs 230 per day. The

Minimum Wages Act, 1948 has stipulated

fixation and enforcement of minimum

wages in the country. As per the act the

average monthly wages of a semi skilled

labour works out to be approximately Rs

9,000. The existing earnings of an IPT

driver are much below this level and does

not provide him with enough resources to

provide education for his children or ensure

good health of his family.

If the minimum wages are considered as

the saving after excluding the expenses

incurred by a driver(which includes rental to

be paid, operation and maintenance cost and

other miscellaneous cost), the fare per

kilometer works out to be as follows (Table

6):

Table 6 Expected Fare per Kilometer (based

on 2013-2014 prices)

Sl.No Heads Autos

(General

Service)

(Rs)

Tempos/

Vikrams

/ Tata

Magic

(Shared

Srvice)

(Rs)

1 Total

Earnings

24,000 35,300

2 Cost per

kilometer

5.8 1.7

3 Saving per

kilometer

3.5 0.6

4 Fare per

kilometer

9.2 2.3

The following assumptions have been taken

Average number of working days in a

month are taken as 26

Fuel cost per day for Autos is Rs 250 and

for Tempos it is 300 (from primary

survey)

Average 100 kilometers are operated per

day as per survey

Average rent per day for autos is Rs 250

and for tempos it is Rs 650 as per survey.

Average maintenance and cost as per

Table 3 is Rs 1,700 for autos and Rs

1,300 for based on the survey

Average miscellaneous expenses as per

Table 3 are Rs 300.

In case financing is available for the IPT

drivers under the SPV, the expected fare per

kilometer would be reduced by a substantial

amount Rs 1.5 / km for auto rickshaws and

Rs. 0.8 / km for tempos (Table 7).


109

Table 7 Expected Fare per Kilometer

Sl.No. Heads Autos

(General

Service)

(Rs)

Tempos/

Vikrams/

Tata

Magic

(Shared

Srvice)

(Rs)

1 Total

Earnings

20,000 24,000

2 Cost per

kilometer

4.2 0.9

3 Saving

per

kilometer

3.5 0.6

4 Fare per

kilometer

7.7 1.5

The following assumptions have been taken:

The capital cost of an auto is Rs 1.3 lakh

an of tempo/ Tata Magic is Rs 3 lakh

(based on the market prices)

The repayment period for the loan is

taken as 7 years

The daily expenditure towards payment

of EMI is Rs 95 for autos and Rs 215 in

case of Tata Magic

The Average vehicle occupancy is 6

passengers

The existing fare across most of the

cities, except in metropolitan areas, is less

than the expected fare given in Table 6&7.

It therefore calls for revision of fare in most

of the cities. Also, this fare would need to be

reviewed periodically – say quarterly or

biannually to reflect the changes in fuel

price or wage rate.

Based on the above calculations, the

proposed fare fixation formula for autos is

as follows:

𝐹𝑎𝑟𝑒𝑝𝑒𝑟𝑘𝑚 = 2 ∗ [(0.07% ∗ 𝐶𝑉)

+ (𝐾𝑀 ∗𝐹𝐶

𝐹𝐸) + (0.26

∗ 𝑊)]

The proposed fare fixation formula for

tempos/ Tata Magic is as follows:

𝐹𝑎𝑟𝑒𝑝𝑒𝑟𝑘𝑚 = 1.6/𝑃 ∗ [(0.07% ∗ 𝐶𝑉)

+ (𝐾𝑀 ∗𝐹𝐶

𝐹𝐸) + (0.21

∗ 𝑊)]

Where:

CV = Capital cost of the vehicle

KM = Average Kilometers operated per day

FC = Cost of Fuel

FE = Fuel Efficiency the vehicle

P = The average occupancy the vehicle

W = minimum daily wages as per Minimum

Wages Act, 1948

For the provision of social benefits to

drivers the government/ SPV / private

bodies/NGOS/ unions of IPT vehicles must

spread awareness to the drivers and may be

advised to become a part of the schemes

provided by the Government like

Janta Personal Accident Insurance

promoted by New India Insurance

Company, free medical check-upsat

various government hospitals for the

drivers and their families, Swavalamban

Pension Scheme provided by Pension

Fund Regulatory and Development

Authority and Sarva Siksha Abhiyan,

free education for school children. Adult

education shall also be promoted for the

drivers though / Government/NGOs/

private bodies/ SPV for example.

6.0 CONCLUSION

Use of IPT is extensively in Indian cities.

They are not only operating in small and

medium sizes cities, but even in popular

large cities as they are playing an important

role in providing mobility at low cost to a

large section of the society. Where in small

cities and medium cities they are acting as

the public transport system, in larger cities

they act as a feeder service to the existing


110

public transport. However, in spite of the

important role that it plays there are various

issues/ challenges related to the sector. In

order to solve most of the problems faced by

this sector organizing the IPT under the

umbrella of an existing SPV or setting up a

new SPV (in case an existing SPV is not

available) is most suitable as it would

organize the existing/ new IPT services

under the umbrella organization, without

any heavy financial burden on the

government.

REFERENCES

1. Sharma, R.S., S.Jain, and

K.Singh.Growth rate of motor vehicles in

India:Impact of Demographic and

Econmic Development. Journal of

Economic and Social Studies, 2011,

pp.137-150.

2. Government of India, Transport

Research Wing Ministry of Road

Transport and Highways, Road

Transport Yearbook 2011-12, 2013.

3. National Transport Development Policy

Committee. India Transport Report,

Moving India to 2032.Publication

Rouledge Taylor and Francis Group,

2014.

4. Government of India. Planning

Commission, Recommendations of

Working Group on Urban Transport for

12th Five Year Plan, 2011.

http://planningcommission.gov.in/aboutu

s/committee/wrkgrp12/hud/wg_%20urba

n%20Transport.pdf

5. Business plan for operations of feeder

services DMRC, UMTC (2009).

6. Government of India, National Urban

Transport Policy,2006.

7. Government of India. The Central Motor

Vehicle Rules, 1989.

http://www.tn.gov.in/sta/Cmvr1989.pdf

8. Government of NCT of Delhi. Transport

Department. Delhi Motor Vehicles Rules,

1993 (As amended).

http://transport.delhigovt.nic.in/pdf/DMV

R.pdf

9. Government of Karnataka. The

Karnataka Motor Vehicle Rules, 1989.

http://rto.kar.nic.in/Revised%20M.V.%2

0VEHICLES%20RULES%20Corrected.p

df

http://planningcommission.gov.in/aboutus/committee/wrkgrp12/hud/wg_%20urban%20Transport.pdf



http://www.tn.gov.in/sta/Cmvr1989.pdf

http://transport.delhigovt.nic.in/pdf/DMVR.pdf

http://transport.delhigovt.nic.in/pdf/DMVR.pdf

http://rto.kar.nic.in/Revised%20M.V.%20VEHICLES%20RULES%20Corrected.pdf



INSTITUTE OF URBAN TRANSPORT (INDIA)

Institute of urban transport (India) New Delhi, a professional non-profit organization was setup in

1997 under the purview of Ministry of Urban Development (MoUD), Government of India (GoI). The

Secretary, MoUD, GoI, is the ex-officio President of Institute and its Governing Council has members from central ministries, states government and various premier organizations connected with

Urban Transport. It has 13 chapters aligned to various institutions of repute.

OBJECTIVE The objective of the institute is to promote, encourage and coordinate the state of art of sustainable

urban transport. Accordingly, the role of IUT has been identified as a ‘National Level Facility’ for

continuous advice and guidance on the principles of sustainable urban transport in National Urban

Transport Policy (NUTP), 2006, National Mission Sustainable Habitat (NMSH) and National Transport Development Policy Committee (NTDPC)

MAIN ACTIVITIES

IUT undertakes a wide range of activities such as research, capacity building, publications, providing professional inputs to Government and non-Government agencies. The activities of IUT are divided

into 5 heads as follows:

Technical Support to Urban Transport Division of MoUD-GoI State and City Government. IUT

has MoUs with Ghaziabad (UP), Bhopal (BCLL), Chandigarh (UT) and the states of Uttrakhand and Bihar for advisory services.

Policy, Research & Projects – IUT has been involved in identifying critical areas needing

research in the field of Urban Transport at National Level and has taken up several projects and

working on various policies as follows:

Evaluation criteria for Urban transport projects Institutional reforms

Area improvement plan

Traffic Management plan Improvement plan of Junctions and roads

Impact assessment of proposed road projects

Design and development of pedestrian and bicycle facilities

Operational plan and ITS plan for city bus service Research study on Service Level Benchmarks (SLB) for urban transport

Working group report of 12th Five year plan and NTDPC

Study to improve and upgrade IPT vehicles and services in India

Training and capacity building –

Preparation of training manuals and toolkits on urban transport

Develop technical program, identify speakers, provide faculty and training material

Training and education of government officials, organizing national and international conferences, seminars, workshops on various urban transport related topics like public transport

(city bus service, bus rapid transit, metro, monorail, light rail transit, etc.), road safety,

institutions, non-motorised transport and governance, etc.

Organize the Urban Mobility India Conference cum exhibition, an annual event held at the national level with participation from India and abroad.

Publication – The institute issues periodic newsletters and journals of Urban transport. In addition

it has also been involved in various publications which are of interest to the nation and many of

them have been launched by MoUD at various occasions.

Information and Library – the institute has a fully functional library with more than 1700 books,

reports and journals related to various aspects of urban transport. Also, managing Knowledge

Management Centre (KMC)

Advisory issued by MoUD quoted that ‘State and city authorities can appoint IUT as Knowledge

Partner for seeking technical support from IUT’. Copy of the advisory available at

http://www.urbanindia.nic.in/programme/ut/Advisory_State_Govt_NLF.pdf

URBAN TRANSPORT

Journal

Volume 13 No.1 September 2014

Contents


Neelakshi Joshi, Prof. R. Shankar and Prof. Dr. Ing Helmut Bott

1

Event Day Effect on Pedestrian Characteristics for CBD Street of Indian Metropolitan

City

Hardik S Sukhadia, Sanjay M Dave, Jiten Shah, Dipak Rathva

14


Prasham Khadaiya and Rajat Rastogi

26

Service Quality Determinants for Public Transport and Use Intention: A Study of

Commuters and Non-Commuters in India

Dr. Vibhuti Tripathi and Gunjan Nema

39

Generic Framework for Estimating Carbon Footprint of Commuting with Public

Transport Modes

Kirti Bhandari, Mukti Advani, Purnima Parida

57

Modeling Mode Choice Behaviour and Estimating Value of Travel Time of Commuters

in Delhi

Minal and Ch.Ravi Sekhar

67


Abdullah Ahmad, Srinath Mahesh and Rajat Rastogi

78

Framework for Development of Advanced Traveler Information System: A Case Study

for Chandigarh City

Bhupendra Singh, Ankit Gupta, Sanjeev Suman

87

Institutional and Financial Strengthening of Intermediate Public Transport Services in

Indian Cities

Anindita Ghosh and Kanika Kalra

96

journals/urban...2 urban transport journal editorial board dr. sanjay gupta (chairman) school of...

Documents