queue evolutions queue evolution is one of the most important factors in design of intersection...
TRANSCRIPT
Queue evolutionsQueue evolution is one of the most important
factors in design of intersection signals. The evaluation compares the model-estimated and the field-observed queue evolutions. Lane 1 is scooter-prohibited, whereas lane 3 serves both scooters and cars.
With an approach of 190M long, it is observed that scooters arrive at the stop-line before cars with a leading time of 8 seconds, which reflects the real-world scooter-vehicle mixed traffic scenario.
Lane ChoiceThe field observation reflects the following
patterns:i. Cars tend to stay on the same travel lane, but
will perform lane changes to merge into the lane that serves their turning needs.
ii. Scooters, with high mobility, tend to choose the lane with fewer vehicles to advance to the downstream stop line.
Travel Time EstimationCars will decelerate and join the end-of-queue;
scooters may decide to filter through the stop queues and not reduce their speed to zero.
Field DataThe scooter-mixed traffic data was collected from the
northbound approaches of an 8-lane major arterial, Xinsheng South Road, in Taipei, Taiwan.
Traffic volume during data collection is about 9,000 vehicles per hour in the target direction; among all vehicle flows, 83% are scooters, 15% are cars, and 2% are buses.
Scooter Discharge RatesThe discharge rates for cars and scooters under non-
mixed traffic condition are set to be 1,800 and 14,000 vphpl, respectively.
An Empirical Study of the Scooter-Vehicle Mixed Traffic Propagation on Urban Arterialsby Chien-Lun Lan and Gang-Len Chang
University of Maryland, College Park
AbstractAbstract
Scooters are one of the primary transportation modes in many developing countries, but design guidelines and software for arterial signals accommodating heavy scooter-vehicle mixed flows are still in their infancy.
Traffic professionals often have no choice but to apply existing models that cannot address scooters’ complex maneuvers.
This study conducts field observations of the mixed traffic flow from their discharging to the formation of stop queues.
Based on the statistical analysis results, this study develops a series of formulations to describe the behavior of mixed traffic flows.
The developed models are evaluated and confirmed to be reliable to serve as the basis for designing arterial control plans.
Research BackgroundResearch Background
Field Observations To study scooter-vehicle mixed traffic, this
research selects two consecutive arterial approaches for field observation.
By using camcorders mounted on a high-rise building, the vehicle trajectories are retrieved on a frame-by-frame basis.
Research Objectives This research tries to understand the
fundamental properties of scooter-vehicle mixed traffic flows, focusing on empirical investigation of their macroscopic behavior.
This study further formulates all statistically validated empirical relations with a series of traffic models.
ConclusionsConclusions
This paper presents the results of empirical investigation with respect to the evolution of scooter-mixed traffic flows from discharging, propagation, to the formation of queues at the downstream intersection.
A series of concise equations are proposed to describe the complex propagation process over an arterial link.
The numerical evaluations of the proposed models with field observations have offered insights into the properties of mixed traffic flows, and can serve as the basis for development of an arterial signal progression model.
The Proposed ModelThe Proposed Model
Discharge ProcessIt is noticeable that scooters often form several
queue lines in a lane while cars can only form a single queue. The field survey reveals that the discharge rate for such mixed flows may vary with the number of scooters in the stop queue.
Acceleration/DecelerationThe average acceleration/deceleration
rates between cars and scooters from the field survey are different.
Comparing to cars• Scooters have higher initial acceleration
rate• Scooters brake later but harder
Merging into Stop QueueAs vehicles approach the stop line and observe
significant differences in queue length among the neighboring lanes, they may perform lane-changing to merge into the lane with shorter queue.
Scooter Filtering ProcessScooters, due to their small physical sizes, tend to
advance between vehicle queue-lines to reach available downstream spaces. It is observed from the field data that scooters’ filtering speed decreases as the mixed flow density increases.
EvaluationsEvaluations
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%0
500
1000
1500
2000
2500
020004000600080001000012000140001600018000
Discharge Rates underDifferent Scooter-Mixed Ratios
Car ScooterSpace Occupied by Scooters in the Stop Queue (%)
Car D
isch
arge
Rat
e (c
ar/s
)
Scoo
ter D
isch
arge
Rat
e (s
coot
er/s
)
0
0.5
1
1.5
2
2.5
3
3.5
0 10 20 30 40 50 60 70 80
Acce
lera
tion
rate
(m/s
2 )
Distance from stop line
Acceleration Rate
Car
Scooter
0
0.5
1
1.5
2
2.5
0 10 20 30 40 50 60 70 80
Dec
. rat
e (m
/s2 )
Distance to the end-of-queue
Deceleration Rate
Car Scooter
60.0% 65.0% 70.0% 75.0% 80.0% 85.0% 90.0% 95.0% 100.0%1
1.5
2
2.5
3
3.5
Filtering Speed under Different Queue Conditions
Occupied space in the stop queue segment(%)
Filt
erin
g Sp
eed
(m/s
)
0%
20%
40%
60%
80%
100%
-4 -2 0 2 4 6 8 10 12
Lane
cha
ngin
g pr
obab
ility
(%)
Queue length difference (vehicles)
Lane Changing Probabilities under Different Queue Length Differences
Logit Model
Real Data
2 0.89
256
R
N samples
Time
Trav
el S
peed
Propagation distance l lL t
,ˆacc lv t
v
,ˆ ffs lv t ,ˆdec lv t
Time
Trav
el S
peed
v
FuPropagation
Filtering l lL t
l t
Speed trajectory of vehicles joining the end-of-queue with and without filtering behavior
0 10 20 30 40 50 60 70 800
10
20
30
40
50
60
Car Speed
Distance to the end-of-queue (m)
Spee
d (k
ph)
0 10 20 30 40 50 60 70 800
10
20
30
40
50
60
70
Scooter Speed
Distance to the end-of-queue (m)
Spee
d (k
ph)
Speed of vehicles during deceleration process
4% 19% 23% 47% 57% 67% 86%
REAL 626 2800 4168 6814 6267 8585 12462
ESTIMATED 596 2710 3231 6553 7913 9333 12000
Difference -5% -3% -22% -4% 26% 9% -4%
-50%
-30%
-10%
10%
30%
50%
0
2000
4000
6000
8000
10000
12000
14000
Scoo
ter d
isch
arge
rate
(vph
)
Scooter occupied space (%)
Scooter Discharge Rate
0
5
10
15
20
25
30
35
40
45
50
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84
Que
ue L
engt
h (m
eter
s)
Time after red phase start (seconds)
Lane 1 (car-only)
Estimated Real
0
5
10
15
20
25
30
35
40
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84
Que
ue L
engt
h (m
eter
s)
Time after red phase start (seconds)
Lane 3 (scooter-mixed)
Estimated Real
0
0.5
1
1.5
2
2.5
3
0
1
2
3
4
5
6
7
8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Arriv
al R
ate
(veh
/sec
ond)
Que
ue le
ngth
(met
ers)
Time (seconds)
Queue Length & Arrival Rate of Lane 3
Arriving Scooters Arriving Cars
Estimated Real
Lane 1
Lane 2
Lane 3
2vI t
1vI t 1
,v mn t
2,v mn t
2,3,v mE t
2 2,v m vA t t
Sub-lane 1
Sub-lane 2
Sub-lane 3
2 2ˆm vt t
2 212 2
1 2 21
m
t tt t
u t t
2 2,v m vQ t t
1 1,v m vA t t
2v t
Lane ChoiceAcceleration
& DecelerationMerging into Stop Queues Stop Queues
Scooter Filteringthrough Stop Queues
Upstream Arrivals
2,v mX t
2,11,mx t
Acceleration/deceleration rates of vehicles
Sub-lane 1
Sub-lane 2
Sub-lane 3
ˆ l t l
ll
tt
u t
,lv mQ t
Filtering Speed lu t
Filtering Distance l t
Car Queue Line