application of analytic hierarchy process for …docs.trb.org/prp/12-0471.pdfregmi and hanaoka 1 1 2...

Regmi and Hanaoka 1

1 Application of Analytic Hierarchy Process for Location Analysis of Logistics 2

Centers in Laos 3 4 Madan B. REGMI (Corresponding author) 5 Department of International Development Engineering 6 Tokyo Institute of Technology 7 Tokyo 152-8550, Japan, 8 Telephone/Fax: 03-5734-3468 9 E-mail: H [email protected] 10 11 Shinya HANAOKA 12 Associate Professor 13 Department of International Development Engineering 14 Tokyo Institute of Technology 15 Tokyo, 152-8550, JAPAN 16 Telephone/Fax: 03-5734-3468 17 E-mail: [email protected] 18 19

Paper submitted on 26 July 2011 for: 20 Presentation at 91th Annual Transportation Research Board Meeting, January 2012 and 21

publication in the Transportation Research Record: Journal of Transportation Reserch Board 22 23

Words in the textt =4,990 24 Number of Figures (4*250) =1000 25 Numbers of Tables ( 6*250 =1,500 26 Total word count =7,490 27

28 29

TRB 2012 Annual Meeting Original paper submittal - not revised by author.

Regmi and Hanaoka 2

Abstract: There have been many efforts to improve transport connectivity in South-East Asia. 1 Strategic location of integrated logistics centers can facilitate transportation, handling, storage 2 and transshipment of goods moving in international trade. The paper proposes to consider 3 additional factors for evaluation of location of logistics centers in addition to the commonly used 4 economic and environmental assessment tools. The paper utilizes the analytic hierarchy process 5 to analyze development of logistics centers in Laos which aims to transform itself into a “land-6 linked” country from a land-locked country. Location models are developed based on the 7 feedback of related government officials and private sector transport service providers. The 8 results show similarity in final priority ranking of the proposed logistics centers, However, some 9 difference on the weights of the evaluation criteria were noted for the public and private sector 10 model. Finally, some policy considerations are offered for development of freight transport and 11 logistics. 12 13 Key words: Logistics Centers, Analytic Hierarchy Process, Freight Transport, Laos 14

15


Regmi and Hanaoka 3

INTRODUCTION 1 2 There have been many efforts to improve transport connectivity in South-East Asia that include 3 initiatives like Greater Mekong Subregion1 (GMS) (1), Association of Southeast Asian Nations 4 (ASEAN) connectivity (2) and ASEAN transport strategic master plan (3). In order to foster 5 seamless flow of goods in international trade the concept of integrated logistics is emerging that 6 provides door-to-door services for freight. The concept embodies integration of various modes of 7 transport, facilities and provision of transport related services at one location. The strategic 8 objective of GMS programme is to complete transport corridors linking the subregion through 9 improved logistics opportunities (1). GMS region has received much foreign direct investment to 10 improve transport infrastructure (4). 11 12

Thus location of logistics facilities is an important consideration that would serve existing 13 trade flow and consumption and production pattern. It is even more important for land-locked 14 countries to develop such facilities in inland area that improve connectivity to ports, markets, 15 production centers and other logistics centers. One of the focuses of the World Bank transport 16 strategy is development of multimodal transport through corridor and logistics development (5). 17 18

Laos, the only land-locked country in South-East Asia, has national logistics strategy to 19 be a logistics hub for GMS and a land-linked country. More than 90% of Laos transit trade goes 20 through Nongkhai, Thailand (6). Therefore, it is important to develop logistics facilities to 21 promote export and import to/from Thailand as well as to facilitate trade among GMS countries. 22 23

Further, development of intermodal transport and integrated logistics centers at strategic 24 locations could be one of the ways to promote environmental sustainability. These facilities by 25 integrating various modes of transport could encourage modal shift to greener modes of transport 26 and reduce less-than-truck-loads trips through provision of consolidation and distribution 27 services. Therefore, in addition to utilization of popular economic and environmental assessment 28 tools other important factors need to be considered to evaluate locations of logistics centers. 29 30

In this context, the paper aims to analyze the facility location problem of logistics centers 31 considering the case of Laos. We used the analytic hierarchy process (AHP) (7) one of the multi-32 criteria decision analysis (MCDA) approach. We developed AHP location models based on the 33 feedback of public and private sector stakeholders in order to prioritize and rank alternative 34 locations of logistics centers. Finally, some policy thoughts are offered for development of 35 logistics and freight transport in Laos. 36 37 2. LITERATURE REVIEW 38 39 Logistics centers are referred as geographical groupings of independent companies or bodies 40 which are dealing with freight transport and with accompanying services including at least 41 terminal (8). The areas of logistics centers, freight villages and intermodal freight hubs, 42 intermodal transport have been extensively researched in Europe and developed countries (9, 43 10). The Marco Polo project aims to increase the share of rail freight in Europe (11). 44 Environmental benefits of intermodal transport in Asian context have also been evaluated (12). 45 1 GMS includes China, Cambodia, Laos, Myanmar, Thailand and Vietnam.


Regmi and Hanaoka 4

1 There are many stakeholders and issues that need to be considered while planning and 2

developing logistics centers. Lack of clear policies and institutional arrangements, conflicting 3 interest of stakeholders pose difficulty in selecting location of logistics centers. While some of 4 established economic analysis tools such as cost benefit analysis, net present value, internal rate 5 of return, environmental impact assessment can look at economic, financial and environmentally 6 feasibility. These tools are commonly used to evaluate alternatives and make decisions based on 7 economic parameters. 8 9

The principle criteria for location is to support movement of goods for industry, 10 consumer and market and to minimize total freight ton-km travelled. In addition to those 11 mentioned above, some of the common factors that affect the facility location are its proximity to 12 ports, markets and manufacturing area, connections to other modes of transport, costs of 13 development, operation and transport, potential of encouraging mode shift, CO2 emissions 14 reduction potential, delivery time, security, and existing government policies. However, many of 15 these factors are subjective and cannot be quantified and thus make the facility location problem 16 a complex. 17 18

MCDA has been extensively used to analyze problems that involve many stakeholders’ 19 views and considers many factors including those involving location problems. MCDA was used 20 to evaluate alternative options for environmentally sustainable transport system in Delhi (13) and 21 assessment of transport projects (14). AHP and Analytic Network Process (ANP) were used to 22 study and evaluate potential location of logistics centers/inland ports in China (15, 16), freight 23 village in Greece (17) and intermodal transshipment and hub ports in Turkey (18, 19, 20). AHP 24 was used to analyze location of wind observation station (21), to measure performance of 25 intermodal transport in Thailand (22), infrastructure management (23), appraisal of transport 26 projects in Korea (24), and prioritization of pavement maintenance activities (25, 26). AHP and 27 ANP were applied for environmental impacts assessment and evaluation of transport policies 28 (27, 28, 29). 29

30 Facility location of inland ports in the United States was analyzed with objective of 31

minimizing vehicle-miles travelled and promoting regional freight transport (30). An agent based 32 model was developed to analyze location of inland ports and hubs in Australia (31). AHP was 33 used to prioritize transport projects (32), to select railway station site in Iran (33) and to select 34 port in Nigeria (34). 35 36

Therefore, from literature it can be seen that the location problem can be analyzed by 37 utilizing MCDA approach. The decision problem involves consideration of various factors and 38 views of various stakeholders that include government decision makers as well private sector 39 transport service providers. AHP one of the MCDA methodologies (35) can incorporate 40 feedback of various stakeholders and utilizes this in prioritizing and ranking alternatives. The 41 subjective as well as objective criteria and sub-criteria that are relevant to the problem at hand 42 can be incorporated in AHP. AHP has been widely used for planning and resource allocation, 43 prioritization of alternatives- which is the problem at hand (36). As many criteria and preferences 44 could be conflicting- a set of solutions with ranking could be presented to the policy makers to 45 assist in the decision making process. 46


Regmi and Hanaoka 5

1 3. LOGISTICS CENTERS IN LAOS 2

3 Laos, a land-locked country, is centrally located in the GMS within South-East Asia. It uses sea 4 ports in Thailand and Vietnam for export and import of goods. One of the routes of the 5 Singapore-Kunming rail project as well as the East-West GMS corridor linking Vietnam to 6 Thailand and Myanmar also passes through Laos. Thus it has a potential to offer much needed 7 land transport connectivity to China and Vietnam. Laos aspire to be seen as a “land-linked” 8 rather than “land-locked’ country with a vision to be a regional logistics hub in GMS. 9 10

Laos total import and export was US $ 1,092 million (equivalent to 1.67 million tons) and 11 US $ 2,430 million (equivalent to 1.963 million tons) in 2009 (37). Figure 1 shows the steady 12 growth of volume of foreign trade with its neighbors (38). It shows that more than 80% of its 13 trade is with Thailand, followed by Vietnam with 13% and China with 8%. Interestingly the 14 volume of trade with southern neighbour Cambodia is very low. Therefore, transport connection 15 to and from Thailand is very important for Laos. Thailand is preferred to Vietnam as main transit 16 route for Laos’s imports and exports (39). 17 18 19

20 21

FIGURE 1 Laos foreign trade volume. 22 23

Laos is well connected by roads with Thailand. Two bridges across Mekong have already 24 been constructed at Thanalaeng and Savannakhet. Currently construction of other two Mekong 25 bridges are progressing one at Huayxai and Thakhek. National Highway-13 (Asian Highway-12) 26


Regmi and Hanaoka 6

is the backbone of road transport. In addition to the roads there is one railway connection to 1 Thanalaeng from Thailand. There are also plans to improve railway connectivity and feasibility 2 studies of connection from Boten, China-Vientiane, Vienatine-Thakhek-Vietnam border are 3 progressing (34). During data collection in Laos it was learnt the feasibility study of Boten-4 Vientiane was currently being reviewed. It was estimated that the construction cost would be 5 high due to the mountainous topography and the need construct many high bridges and tunnels. 6 Vientiane–Thakhek rail route is also in pipeline for development. It is expected that Laos railway 7 network will expand in near future. The Mekong River is also used for water transport. The 8 current mode share is 80% by road and 20% by water transport. ADB has estimated the freight 9 demand for GMS 6.3 million tons for 2014 and 23.8 million tons for 2025 (40). 10 11

Laos logistics performance index (LPI) is 2.46 which is better than its neighbours 12 Myanmar (2.33) and Cambodia (2.37) while China (3.49), Thailand (3.29) and Vietnam (2.96) 13 have higher LPI. If we compare with other landlocked countries it is better than that of Nepal 14 (2.2), Mongolia (2.25), Bhutan (2.38) and Afghanistan (2.24). However Kazakhstan (2.83 have 15 higher LPI than Laos (41). 16 17

18 19

FIGURE 2 Study area and location of logistics centers in Laos. 20 21

Improvement of logistics system and transport connectivity would be the core activity to 22 materialize the vision of being a land-linked country. Laos current development plan forecast a 23 7% growth of freight that will reach 32 million tons and 2.2 billion ton-Km by 2015 and focus 24 on development of special economic zones already established in Boten, Bokeo and Savannakhet 25 (42). Thailand also has railway capacity enhancement plan that includes double tracking to 26 Nongkhai, track rehabilitation and development of inland container depots in Nongkhai and 27


Regmi and Hanaoka 7

Ubon Ratchathani near border with Laos (43). Laos thus can use these facilities as well as the Lat 1 Krabang ICD near Bangkok which is connected by railway to Laem Chabang port. 2 3

Laos national strategic logistics master plan (44) has proposed to develop three 4 international logistics parks at Thanalaeng (Vientiane), Savannakhet and Luangnamtha and three 5 regional logistics parks at Luangparbang, Thakhek and Pakse (Champasak) which are shown in 6 Figure 2. Due to the importance of international trade, potential freight volume and transport 7 connectivity with Thailand as well as GMS four logistics centers namely Thanalaeng, Thakhek, 8 Savannakhet and Pakse are considered as cases in the current study. These four alternative 9 locations have potential to serve international and transit as well as domestic freight. The salient 10 features of these logistics centers are shown in table 1. 11 12

TABLE 1 Salient Features of Proposed Logistics Centers in Laos (44) 13 14

Freight volume Design Area Costs Functions and roles Tons for 2015 Tons for 2025 sq. m Mi US$

Thanalaeng 426,500 1,413,060 295,000 31.1 Road, rail and waterway, import and export to Thailand, limited transit to China, consolidation, transshipment and distribution

Thakhek NA 79,000 23,300 3.7 Road and waterway, import, export and transit to/from Thailand and Vietnam, consolidation, transshipment and distribution, bridge to Thailand near completion

Savannakhet 36,000 151,000 51,100 4.56 Road and waterway, import, export and transit to/from Thailand and Vietnam through GMS East-West Corridor, consolidation, transshipment and distribution,

Pakse 106,400 296,000 115,900 14.77 Road and waterway, import, and export to/from Thailand, transit to Cambodia, consolidation, transshipment and distribution, Storage and distribution of agricultural products

15 Despite having logistics development plan Laos is finding it difficult to develop these 16

logistics facilities. Prioritization of logistics facilities would help policy makers to take decision 17 and mobilize resources according to the priority and gradually improve overall logistics system. 18 19


Regmi and Hanaoka 8

4. METHODS, DATA AND ANALYSIS 1 2 AHP can consider both quantitative and qualitative criteria and derive relative weights and 3 prioritize alternate locations based on evaluation criteria selected. Using a systematic hierarchy 4 structure, complex estimation criteria can be clearly and precisely represented (45). The AHP 5 involves the following process: 6

(i) Hierarch development- defining goal, evaluation criteria and sub-criteria and 7 alternatives; 8

(ii) Pair-wise comparison of criteria, sub-criteria and alternatives; 9 (iii) Weight derivation of criteria, sub-criteria and alternatives; 10 (iv) Synthesis- derivation of priorities; 11 (v) Consistency check. 12

13 14 4.1 AHP Model 15 16 The facility location problem is decomposed into four levels of hierarchy consisting of (i) one 17 goal; (ii) five criteria; (iii) 12 sub-criteria; and (iv) four alternatives. Relevant criteria and sub-18 criteria are selected after reviewing the elements considered by earlier studies (15, 16, 17, 18, 19, 19 20, 33, 46, 47, 48). The criteria and sub-criteria were further refined after consultation with the 20 related officials in Laos. The four alternatives selected are the location of real logistics centers 21 proposed in the recent logistics study (44). The selected criteria and sub-criteria and the AHP 22 model are shown table 2 and figure 2. This decomposition of decision problem makes it possible 23 to judge the importance of the elements in a given level with respect to other elements in the 24 same level. The number of main criteria has been limited to five in order to limit numbers of 25 questions for pair-wise comparisons and to increase probability of receiving feedback on the 26 survey. AHP assumes the criteria and sub-criteria should be independent of each other. 27 28

TABLE 2 Goal, Criteria, Sub-criteria and Alternatives of AHP Location Model 29 Goal Criteria Sub-criteria Alternatives Prioritize and rank location of logistics centers

Development and operation costs (C1)

Land acquisition costs (C11) Construction costs (C12) Transportation costs (C13)

Thanalaeng (A1) Savannakhet (A2) Thakhek (A3) Pakse (A4)

Time (C2) Total transport time from seaport (C21)

Intermodal transport connectivity (C3)

Highways (C31) Railways (C32) Inland waterways (C33) Seaports (C34)

Environmental impacts (C4)

Impacts from construction (C41) Impacts from transport

operation(C42) Regional economic development (C5)

Proximity to market, production centers and consumers (C51)

Government polices to develop


Regmi and Hanaoka 9

special economic zone or free trade area (C52)

Freight demand (C53) 1

2 FIGURE 3 AHP location analysis model of logistics center. 3

4 4.2 Weight Derivation 5

Relative weights of decision criteria and sub-criteria are derived by making pair-wise 6 comparison for all the selected criteria at each level. The main objective of this process is to 7 arrive at priority of the alternative locations based on the criteria and sub-criteria selected. The 8 respondent is asked to indicate value judgment/preference in a 1 to 9 point Saaty’s scale while 9 making pair-wise comparison. For quantitative criteria the corresponding values are estimated 10 based on the available data from survey and the value assigned from pair-wise comparison are 11 replaced. It may not be always possible to measure and present these factors as numerical values. 12 Rather these factors are expressed as value judgment, preferred options based on criteria. 13 14 Individual judgments are combined to arrive at the group judgment typically by taking 15 the geometric mean method (GMM) of the individual judgment (49). The aggregate weight of 16 individual judgments is calculated by the following equation. 17 18 19

nji

njijijijiji aaaaaa /1

,,4

,3

,2

,1

, )*........................***(= (1) 20


Regmi and Hanaoka 10

1 A judgmental matrix, A= (ai,j) such that ai,i=1, ai,j =1/ aj,i for all i, j≤n, is developed 2 based on pair-wise comparison of priorities of criteria (50). 3 4

⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢

⎣

⎡

=

1.../1/1...........

...1/1

...1

21

212

112

nn

n

n

aa

aaaa

A (2) 5

6 Where, 7 n number of alternatives, aij represents pair-wise comparison rating between the element 8 i and j of a criteria with respect to goal being considered 9 10 Then priorities of the criteria is estimated by calculating the principal eigenvector W of 11 the matrix A, 12 13 14

WWA ** maxλ= (3) 15

16 Where, λmax is principle eigenvalue, when vector W is normalized, it becomes vector of 17 priorities of criteria of one level with respect to the upper level. 18 19 20

jkkiji aaa ,,, *= ∀i,j,k. (4) 21 22

23 After analyzing group judgment, the weights are derived by using by eigenvector method 24 (linear algebra). 25 26

j

iji w

wa =, (for i, j,= 1, 2,......n) (5) 27

28 29 The consistency index (CI) is calculated using equation (6). 30 31 32



)1()( max

−−

=n

nCI

λ

(6) 1

2 CI should be below 0.2 for the survey to be accepted, if CI >0.2 indicates lack of 3 consistency and warrants repeat of survey. 4 5 The Consistent Ratio (CR) is calculated using equation (7), generally CR ≤ 0.1. 6

7

RICICR =

(7) 8 9 Where, RI is Random Index 10 11 The measure of consistency enables analysis to review the judgments, modifying them to 12

improve the overall consistency. 13 14 4.3 Data Collection 15 16 Questionnaire survey for pair-wise comparison of criteria, sub-criteria and alternative locations 17 was developed and refined based on the feedback received from researchers and government 18 officials. The questionnaire involved 101 questions for pair-wise comparison and general 19 information on the respondents. 20

21 Contact was maintained with transport and logistics related government officials, Laos 22

International Freight Forwarders’ Association (LIFFA) and Lao National Chambers of 23 Commerce and Industry (LNCCI) to develop a long list of potential respondents. From the long 24 list 50 people were selected for the survey that included 23 government officials, all (22) 25 members of LIFFA and 5 members of LNCCI. 26 27

Prior to the visit to Laos for survey, the questionnaire was sent to all 50 potential 28 respondents. On site data collection was done during 6-12 July 2011 which also involved visit to 29 all four alternative locations and meeting and discussion with freight forwarders, private sector 30 representatives, and government officials. Total 21 responses were received with an overall 31 response rate of 42%. The response of private sector was 33% (9 response from 27 requests) 32 compared with government policy makers 52% (12 response from 23 requests). The respondents 33 included government officials representing public works and transport, road, railway, transport, 34 national transport facilitation committee, legal affairs, and officials of special economic zones. 35 Private sector included members of LIFFA and LNCCI. 36 37

During the survey it was felt that questionnaire was too long as it contained 101 questions 38 and took between 20-30 minutes to complete the survey. As a result some of the potential 39 respondents promised to provide feedback but never did. 40 41



Two completed questionnaire were incomplete and were discarded. Most of the responses 1 received were consistent. However, due to large numbers of questions, some questions were left 2 unanswered due to possibly “reply fatigues.” 3 4 4.4 Analysis 5 6 Individual survey responses were tabulated in two groups and analyzed formulating government 7 decision makers (GDM) and private sector freight forwarder (PSFF) models. A third model was 8 created combining (COMB) both of the above groups. Table 3 shows the derived weights of five 9 evaluation criteria and table 4 shows derived weight of sub-criteria. The consistency ration (CR) 10 is also shown in the table 3 and 4 which in all case is less than 0.1 indicating consistency of the 11 pair-wise comparison and weight derivation. 12 13

TABLE 3 Derived Weights of Evaluation Criteria 14 Derived weights

Goal Criteria GDM (C.R=0.01)

PSFF (C.R=0.08)

COMB (C.R=0.08)

Prioritize location of logistics center

Development and operation costs (C1) 0.240 0.142 0.220 Time (C2) 0.129 0.131 0.215 Intermodal transport connectivity (C3) 0.230 0.150 0.199 Environmental impacts (C4) 0.170 0.312 0.176 Regional economic development (C5) 0.231 0.265 0.190

15 The above results show that GDM assign high priority to cost and regional economic 16

development, interestingly PSFF assign priority to environmental impacts and regional economic 17 developed while COMB indicates high priority to time and costs. 18 19

TABLE 4 Derived Weights of Sub-criteria 20 Priority weights

Criteria Sub-criteria GDM PSFF COMB

Development and operation costs (C1)

Land acquisition costs (C11) Construction costs (C12) Transportation costs (C13)

0.280 0.353 0.367

(C.R=0.02)

0.266 0.169 0.566

(C.R=0.01)

0.276 0.282 0.442

(C.R=0.01) Time (C2) Total transport time from seaport

(C21) 1 1 1

Intermodal transport connectivity (C3)

Highways (C31) Railways (C32) C33 Inland waterways Seaports (C34)

0.423 0.338 0.149 0.090

(C.R=0.01)

0.281 0.358 0.208 0.153

(C.R=0.08)

0.356 0.355 0.172 0.117

(C.R=0.02) Environmental impacts (C4)

Impacts from construction (C41) Impacts from transport operation (C42)

0.545 0.455

(C.R=0.0)

0.583 0.417

(C.R=0.0)

0.524 0.476

(C.R=0.0) Regional economic development (C5)

Proximity to market, production centers and consumers (C51) Government polices to develop special economic zone or free trade area nearby (C52)

0.425

0.386

0.222

0.505

0.340

0.438



Freight demand (C53) 0.190 (C.R=0.03)

0.273 (C.R=0.02)

0.222 (C.R=0.02)

1 Construction cost (C12), transportation time (C21), proximity to market and production 2

centers (C51) and freight demand (C53) are tangible criteria which can be measured. For land 3 acquisition cost (C11) comparison of land cost at different location is used. Sub-criteria 4 transportation cost (C13) is also measurable but due to sensitivity of the transportation price 5 among freight forwarders and transport operators it could not be collected and the rating 6 provided by the respondents was used instead. The data for these criteria and sub-criteria was 7 collected during the survey and secondary sources. The table 5 shows the tangible data collected 8 and their normalized values which were used for analysis. Except for ‘freight demand’ higher the 9 value of element lower the priority. For example, higher construction cost of an alternative 10 would encourage decision makers to consider other alternative locations having lower 11 construction costs. 12 13

TABLE 5 Estimated Values of Tangible Criteria and Sub-criteria 14 15

Alternatives Land acqu. Cost (C11)

Construction cost (C12)

Transport time (C13)

Proximity to market (C51)

Freight demand (C53)

Ratio Norm.

M US$ Norm. Hours Norm. Km Norm. Tons Norm.

A1 1 0.192 32 0.053 9 0.233 21 0.328 4032000 0.533 A2 0.65 0.295 3.7 0.459 9 0.233 24 0.287 1150000 0.152 A3 0.8 0.240 4.56 0.373 9 0.233 28 0.246 1577000 0.209 A4 0.7 0.274 14.77 0.115 7 0.300 50 0.138 803000 0.106

16 The analysis was done using distributive mode and overall CR of the GDM model is 17

0.02, PSFF model 0.05 and that of COMB model is 0.04. This indicates overall consistency of 18 data and analysis and reliability of the outcome. 19 20

Figure 4 shows final derived weight of four alternative locations using three different 21 AHP models. Detail weight derivation and ranking for three different AHP models is shown in 22 table 6. The results and outputs of three location models ranked Thanalaeng logistic center as the 23 highest priority location followed by Savannakhet and Pakse. This result output is in line with 24 the view of policy makers and freight forwarders. The result highlights the importance of 25 Thanalaeng because its railway connectivity, proximity to Vientiane capital and expected freight 26 volume. Thakhek scored lowest rank in all three models even though it is a shortest route to 27 connect Vietnam. It could be because the bridge over Mekong River connecting Laos and 28 Thailand at Thakhek is still under construction and is planned to be opened on 11 November 29 2011. 30 31



1 FIGURE 4 Final derived weights of alternatives logistics centers. 2

3 4.5 Results and Discussion 4 5 The results further reinforce the priority for Thanalaeng even if there is some dilemma among 6 some policy makers on priority because of the existing policy to promote SSFEZ and 7 Savannakhet logistics center. This argument is further strengthened by the same outcome from 8 GDM and PSFF model. There are some views that Savannakhet is more important for 9 international trade as this is located along GMS East-West Corridor. 10 11

Only logistics infrastructure is not enough to facilitate trade and movement of goods. Its 12 efficient operation and management is the key to support international trade and transport. 13 Development of logistics centers should be driven from real economic potential, implementation 14 of facilitation measures and improvement of road infrastructure (51). Initiating policy or system 15 for single handling of cargoes using ASYCUDA for customs clearance, reducing numbers of 16 documents required for imports and exports, expanding use of single window system, using 17 information and communication technology for cargo/container tracking, initiating freight train 18 services between Bangkok and Vientiane and enhancing use of the Mekong river are some of the 19 measures that would promote freight transport in Laos and contribute towards turning it into a 20 “land-linked” country. 21

22 23

5. CONCLUSIONS 24 25 The location of proposed logistics centers was analyzed using AHP. The analysis showed that 26 Thanalaeng, which is close to Vientiane, ranked high priority followed by Savannakhet, Pakse 27 and Thakhek. These results are consistent with the view of the freight forwarders and 28 government officials. However, some difference in weights of each criteria and sub-criteria were 29 noted among government decision makers and private sector freight forwarders. 30 31

The development of logistics infrastructure is not sufficient in itself it will require more 32 efforts to operate these facilities. More importantly reduction of documents to be handled in 33



dealing with export and import, implementation single window system this eliminating double 1 inspection of cargoes at the border as well and at the logistics centers are some of the policy and 2 system issues that need to considered. There is no “one fits for all policies” these should be 3 implemented by taking a step by step approach. As the railway connectivity to Thailand has now 4 been completed starting regular freight train between Thailand and Laos would help to improve 5 international freight transport and to reduce environmental burden to some extent. By effectively 6 using these logistics facilities as consolidation and distribution centers number of less-than-truck 7 trips could be reduced. These facilities would also help to improve transit transport through Laos 8 and transforming is as a “land-linked” country in South-East Asia. 9


Regm

i and

Han

aoka

16

TAB

LE 6

Fin

al W

eigh

t of A

ltern

ativ

es

D

eriv

atio

n of

wei

ghts

of s

ub-c

rite

ria

Government Decision makers

C

11

C12

C

13

C21

C

31

C32

C

33

C34

C

41

C42

C

51

C52

C

53

Fina

l W

eigh

ts

Ran

k Th

anal

aeng

0.

0128

0.

0045

0.

032

0.03

8 0.

03

0.04

1 0.

01

0.00

4 0.

029

0.02

3 0.

032

0.03

1 0.

023

0.31

1 1

Thak

hek

0.01

98

0.03

90

0.01

4 0.

029

0.01

7 0.

015

0.00

8 0.

005

0.02

2 0.

019

0.02

8 0.

018

0.00

7 0.

241

3 Sa

vann

akhe

t 0.

0161

0.

0317

0.

02

0.03

3 0.

027

0.01

1 0.

008

0.00

7 0.

017

0.01

5 0.

024

0.02

2 0.

009

0.24

1 2

Paks

e 0.

0183

0.

0098

0.

022

0.02

9 0.

023

0.01

0.

008

0.00

5 0.

024

0.02

0.

014

0.01

8 0.

005

0.20

6 4

Private sector

Than

alea

ng

0.00

73

0.00

13

0.03

2 0.

038

0.00

8 0.

022

0.00

8 0.

005

0.05

1 0.

053

0.01

9 0.

014

0.03

9 0.

298

1 Th

akhe

k 0.

0112

0.

0110

0.

013

0.03

0 0.

012

0.01

1 0.

008

0.00

7 0.

048

0.02

2 0.

017

0.03

0.

011

0.23

1 3

Sava

nnak

het

0.00

91

0.00

89

0.02

2 0.

033

0.01

4 0.

012

0.00

7 0.

005

0.03

1 0.

024

0.01

5 0.

058

0.01

5 0.

254

2 Pa

kse

0.01

04

0.00

28

0.01

2 0.

030

0.00

8 0.

009

0.00

8 0.

006

0.05

1 0.

031

0.00

8 0.

032

0.00

8 0.

216

4

Combined

Than

alae

ng

0.01

17

0.00

33

0.03

6 0.

063

0.01

8 0.

034

0.01

0.

005

0.02

8 0.

028

0.02

1 0.

013

0.02

2 0.

293

1 Th

akhe

k 0.

0180

0.

0285

0.

018

0.04

9 0.

016

0.01

4 0.

008

0.00

5 0.

023

0.01

9 0.

019

0.01

7 0.

006

0.24

0 3

Sava

nnak

het

0.01

46

0.02

31

0.02

7 0.

055

0.02

0.

012

0.00

8 0.

009

0.01

7 0.

016

0.01

6 0.

035

0.00

9 0.

261

2 Pa

kse

0.01

67

0.00

71

0.01

6 0.

049

0.01

6 0.

01

0.00

8 0.

004

0.02

5 0.

021

0.00

9 0.

018

0.00

4 0.

204

4



ACKNOWLEDGEMENT 1

The authors acknowledge the support provided by the Ministry of Public Works and Transport, 2 Laos, LIFFA and LNCCI in arranging data collection, meetings with stakeholders and visits to 3 the logistics centers. The views expressed herein are those of the authors and do not necessarily 4 reflect the views of the United Nations. 5

6 REFERENCES 7

8 1. Greater Mekong Subregion. <http://www.adb.org/GMS/> Accessed 24 July 2011. 9 2. ASEAN. Master Plan on ASEAN Connectivity, ASEAN Secretariat, Jakarta, 2010. 10 3. ASEAN. ASEAN Transport Strategic Master Plan, Economic Research Institute for 11

ASEAN and East Asia, Jakarta, 2010. 12 4. Ishida, M. (ed.). Intra-and Inter-City Connectivity in the Mekong Subregion, 2011, IDE-13

JETRO, Bangkok. 14 5. World Bank. Transport for Development, Transport Business Strategy for 2008-2012, 15

Washington, 2008. 16 6. Banomyong, R. Behind Border Issues: Trade Facilitation, Presentation made at Lao 17

PDR: The Trade and Integrated Framework-A Refreshing Workshop, 28-29 June 2004, 18 Vientiane. 19

7. Saaty, T. L. The Analytic Hierarchy Process, McGraw Hill, New York, 1980. 20 8. Economic Commission for Europe. Terminology on Combined Transport, 2001. 21 9. Rodrigue, J.-P., J. Debrie, A. Fremont and E. Gouvernal. Functions and Actors of Inland 22

Ports: European and North American Dynamics. Journal of Transport Geography, 2010, 23 18(4): 519-529. 24

10. Caris, A., C. Macharis, and G. K. Janssens. Planning Problems in Intermodal Freight 25 Transport: Accomplishments and Prospects. Transportation Planning and Technology, 26 2008, 31(3): 277-302. 27

11. Marco Polo-New Ways to Green Horizon. <http://ec.europa.eu/transport/marcopolo/> 28 Accessed 22 July 2011. 29

12. Hanaoka, S. and M. B. Regmi. Promoting Intermodal Freight Transport through the 30 Development of Dry Ports in Asia: An Environmental Perspective. IATSS Research, 31 2011, In Press, Uncorrected Proof. 32

13. Yedla, S. and R. M. Shrestha. Multi-criteria Approach for the Selection of Alternative 33 Options for Environmentally Sustainable Transport System in Delhi. Transportation 34 Research Part A: Policy and Practice, 2003, 37(8): 717-729. 35

14. Tsamboulas, D., G. S. Yiotis, and K. D. Panou. Use of Multicriteria Methods for 36 Assessment of Transport Projects. Journal of Transportation Engineering, 1999, 125: 37 407. 38

15. Wang, C. and J. Wei (2009). Research on the Dry Port Location of Tianjin Port Based on 39 Analytic Network Process, IEEE. 40

16. Wei, J., A. Sun, and J. Zhuang. The Selection of Dry Port Location with the Method of 41 Fuzzy-ANP. Advances in Wireless Networks and Information Systems, 2009, 265-273. 42

17. Kapros, S., K. Panou, and D. Tsamboulas. Multicriteria Approach to the Evaluation of 43 Intermodal Freight Villages. In Transportation Research Record: Journal of the 44 Transportation Research Board, 2005, 1906(1): 56-63. 45



18. Önüt, S. and S. Soner. Transshipment Site Selection Using the AHP and TOPSIS 1 Approaches Under Fuzzy Environment. Waste Management, 2008, 28(9): 1552-1559. 2

19. Onut, S., U. R. Tuzkaya, and E. Torun. Selecting Container Port via a Fuzzy ANP-based 3 Approach: A Case Study in the Marmara Region, Turkey. Transport Policy, 2010. 4

20. Tuzkaya, G., S. Önüt, U. R Tuzkaya, and B. I. Gülsün. An Analytic Network Process 5 Approach for Locating Undesirable Facilities: An Example from Istanbul, Turkey. 6 Journal of Environmental Management, 2008, 88(4): 970-983. 7

21. Aras, H., S. Erdogmus, and E. Koc. Multi-criteria Selection for a Wind Observation 8 Station Location Using Analytic Hierarchy Process. Renewable Energy, 2004, 29(8): 9 1383-1392. 10

22. Kunadhamraks, P. and S. Hanaoka. Multiple Criteria and Fuzzy Based Evaluation of 11 Logistics Performance for Intermodal Transportation. Journal of Advanced 12 Transportation, 2009, 43(2): 123-153. 13

23. Smith, J. and S. Tighe. Analytic Hierarchy Process as a Tool for Infrastructure 14 Management. In Transportation Research Record: Journal of the Transportation 15 Research Board, 2006, 1974(-1): 3-9. 16

24. Lee, Sang-Min. Analytic Hierarchy Approval for Transport Project Appraisal an 17 Application to Korea, 1998, PhD Dissertation, University of Leeds. 18

25. Farhan, J. and T. Fwa. Pavement Maintenance Prioritization Using Analytic Hierarchy 19 Process. In Transportation Research Record: Journal of the Transportation Research 20 Board, 2009, 2093(-1): 12-24. 21

26. Cafiso, S., A. Di Graziano, H. Kareli and J. Odoki. Multicriteria Analysis Method for 22 Pavement Maintenance Management. In Transportation Research Record: Journal of the 23 Transportation Research Board, 2002, 1816(-1): 73-84. 24

27. Islam, R. and T. Saaty. The Analytic Hierarchy Process in the Transportation Sector. In 25 Ehrgott, M., B. Naujoks, et al. (eds) Multiple Criteria Decision Making for Sustainable 26 Energy and Transportation Systems, Springer Berlin Heidelberg, 2010, 634: 79-91. 27

28. Berrittella, M., A. Certa, M. Enea and P. Aito. An Analytic Hierarchy Process for the 28 Evaluation of Transport Policies to Reduce Climate Change Impacts, 2007, Fondazione 29 Eni Enrico Mattei, Milano. 30

29. Brozova, H. and M. R Žika. The AHP and ANP Models for Transport Environmental 31 Impacts Assessment, 2010. 32

30. Rahimi, M., A. Asef-Vaziri, and R. Harrison. An Inland Port Location-Allocation Model 33 for a Regional Intermodal Goods Movement System. Maritime Economics & Logistics, 34 2008, 10: 362-379. 35

31. Sirikijpanichkul, A., K. H. Van Dam, L. Ferreira, and Z. Lukszo. Optimizing the 36 Location of Intermodal Freight Hubs: An Overview of the Agent Based Modelling 37 Approach. Journal of Transportation Systems Engineering and Information Technology, 38 2007, 7(4): 71-81. 39

32. Shelton, J. and M. Medina. Integrated Multiple-Criteria Decision-Making Method to 40 Prioritize Transportation Projects. In Transportation Research Record: Journal of the 41 Transportation Research Board 2174(-1): 51-57. 42

33. Mohajeri, N. and G. R. Amin. Railway Station Site Selection Using Analytical Hierarchy 43 Process and Data Envelopment Analysis. Computers & Industrial Engineering, 2010, 44 59(1): 107-114. 45



34. Ugboma, C., O. Ugboma, and I. C. Ogwude. An Analytic Hierarchy Process (AHP) 1 Approach to Port Selection Decisions Empirical Evidence from Nigerian Ports. Maritime 2 Economics & Logistics, 2006, 8: 251-266. 3

35. Saaty, T. L. That is not the Analytic Hierarchy Process: What the AHP is and What it is 4 not. Journal of Multi-Criteria Decision Analysis, 1997, 6(6): 324-335. 5

36. Saaty, T. L. Transport Planning with Multiple Criteria: The Analytic Hierarchy Process 6 Applications and Progress Review. Journal of advanced transportation, 1995, 29(1): 81-7 126. 8

37. United Nations. Commodity Trade Statistics Database, <http://comtrade.un.org/> 9 Accessed 15 June 2011. 10

38. ADB. Logistics Development Study of the GMS North-South Corridor, Summary, 2008. 11 39. Arnold, J., R. Banomyong, and N. Rithironk. Logistics Development and Trade 12

Facilitation in Lao PDR, 2003, World Bank, Vientiane. 13 40. ADB. Connecting Greater Mekong Subregion Railways, Strategic Framework, 2010, 14

Manila. 15 41. Arvis, J-F., M. A. Mustra, J. Panzer, L. Ojala and T. Naul. Connecting to Compete: Trade 16

Logistics in the Global Economy. World Bank. Washington, DC, 2010. 17 42. Ministry of Planning and Investment. The Seventh National Socio-Economic 18

Development Plan (2011-2015), Draft, (translation from Lao), 2011, Laos. 19 43. Mala, Voravuth. Enhancing Border Connectivity through Intermodal networks: Effective 20

Strategies to Capitalize on Rising Demand for Freight Transport. Presentation made at 21 Modern Railways Conference, Bangkok, 23-25 May 2011. 22

44. Japan International Cooperation Agency, The Comprehensive Study on Logistics System 23 in Lao Peoples’ Democratic Republic, 2011, Final Report, Vientiane. 24

45. Saaty, T. L. Decision Making-The Analytic Hierarchy and Network Processes 25 (AHP/ANP), Journal of System Science and System Engineering, Vol 13, No.1, March 26 2004. 27

46. Guneri, A. F., M. Cengiz, and S. Seker. A Fuzzy ANP Approach to Shipyard Location 28 Selection. Expert Systems with Applications, 2009, 36(4): 7992-7999. 29

47. Yang, C. L., S. P. Chuang, R. H. Huang, and C. C. Tai. Location Selection Based on 30 AHP/ANP Approach, 2009, IEEE. 31

48. Kahraman, C., D. Ruan, and I. Duan. Fuzzy Group Decision-making for Facility 32 Location Selection. Information Sciences, 2003, 157: 135-153. 33

49. Forman, E. and K. Peniwati. Aggregating Individual Judgments and Priorities with 34 Analytic Hierarchy Process. European Journal of Operational Research, 108(1998) 165-35 169. 36

50. Saaty, T. L. Decision-making with the AHP: Why is the Principal Eigenvector 37 Necessary? European Journal of Operational Research, 2003, 145(1): 85-91. 38

51. Nolitha, V. Cities, SEZs and Connectivity in Major Provinces of Laos in M. Ishida (ed.) 39 Intra-and Inter-City Connectivity in the Mekong Subregion, 2011, IDE-JETRO, 40 Bangkok. 41


application of analytic hierarchy process for …docs.trb.org/prp/12-0471.pdfregmi and hanaoka 1 1 2...

Documents