Building Tomorrow’s Society

Bâtir la Société de Demain

Fredericton, Canada
June 13 – June 16, 2018/ Juin 13 – Juin 16, 2018

PERFORMANCE EVALUATION OF A FOUR-LEGGED INTERSECTION

USING MICROSCOPIC SIMULATION MODEL

Orvin, Muntahith1,5, Ivy, Afia2, Hadiuzzaman,Md3 and Ahsan, Hasib4

1
BSc, Military Institute of Science And Technology (MIST), Bangladesh
2
BSc, Military Institute of Science And Technology (MIST), Bangladesh
3
PhD, Professor, Bangladesh University of Engineering And Technology (BUET), Bangladesh
4
PhD, Professor, Bangladesh University of Engineering And Technology (BUET), Bangladesh
5
mmceorvin@gmail.com

Abstract: Traffic congestion is a widely spreading problem throughout the world. The gap between traffic
demand and supply is increasing day by day that makes junctions, a traffic congestion spot. In this study,
a four legged intersection has been evaluated that suffers from high congestion. The traffic within the
study area is mostly heterogeneous with non-lane based flow. A microscopic simulation model has been
used to assess the performance of the intersection. Using VISSIM, the on-field condition is replicated to
observe the behaviour of the prevailing system. Several traffic data are collected from the study area.
Then the simulation model is calibrated with the field data and validated with another day’s field data,
signifies that the developed model clearly mimics the reality. The driving behaviour parameters are tuned
to replicate the lane indiscipline existing in the field. The model is analyzed for several conditions such as
existing two phase, modified two phase and four phase signal control. It is found that travel time for
modified two-phase is more favourable than the existing two-phase signal. Moreover, delay result is more
conducive in modified two-phase system. It is to be noted that the four-phase signal produces maximum
queue length. The signal design modification can be a great solution to improve the existing congestion
condition prevailing in the studied intersection. The strategy used in the model can be incorporated for the
improvement of any intersection adjusting the calibration standards. Increase of efficiency of intersections
suffering congestion will lead to a better urban traffic system.

1 Introduction

On a typical four-leg intersection, one out of two intersecting roads has the higher traffic volume. This
roadway is referred to as the major or arterial road. The second roadway, which services the lower traffic
volume, is referred to as the minor or collector road. When the volume on either road nears capacity,
queues begin to form, raising the potential for unsafe driving manoeuvres. For this reason, improving
safety and operational efficiency at intersections on urban road intersections remain a constant goal
(Naghawi 2014). Intersections are usually considered as the critical points within the network and the
evaluation of their performance provides valuable understanding and useful indication about the
performance of the system (Sunil 2013).

Signalised intersections are one of the key locations in any transportation network, hence improper
planning of the intersection affects the performance and productivity of the whole road network. A
thorough analysis has to be made to utilise the signalised intersections effectively. The analytical
approach has the limitation of underlying assumption of homogeneity, which is having high variations of
traffic characteristics when compared with a mixed traffic condition. Simulation models on the other hand,
follow the dynamic nature of the traffic system and give a continuous view of the state of the traffic system

TR31-1
over time. This characteristic of simulation model is an advantage over analytical models, since it gives
more insight to what is happening in the system in detail. Hence, researchers concentrated on
appropriate modelling technique through simulation, which is emerged as the most powerful and
acceptable solution searching tool (Sheela 2014).

Dhaka is the capital of Bangladesh, one of the mega cities in the world. ‘Banglamotor’ of Dhaka is a busy
four legged intersection during the peak hours. Non-lane based flow predominates and traffic is mixed
with both motorized and NMV’s in the minor road while only motorized vehicle ply on major road due to
the restriction imposed by the road authority. This intersection need to be analyzed for improvement in
terms of various traffic characteristics. In this study this intersection is modeled with microscopic
simulation software. Out of numerous parameters, signal timing and phasing is modified to understand
the traffic properties, how it changes and improves the present condition. The improvement will help the
overall increase in performance of the urban intersection that will eventually help in building a congestion
free society.

Figure 1: Four-legged busy intersection named ‘Banglamotor’ of megacity Dhaka

2 Literature Review

Traffic micro-simulation models are widely used to evaluate the benefits and limitations of traffic operation
alternatives. Traffic simulation models are mainly classified as microscopic and macroscopic. Models that
simulate individual vehicles at small time intervals are termed as microscopic, while models that
aggregate traffic flow are termed as macroscopic. Microscopic simulation has been used for a long time to
simulate project scale cases such as intersection design. Thus computer simulation is a valuable tool for
the analysis and design of complex transportation systems.

Simulation models may be classified as being static or dynamic, deterministic or stochastic and discrete
or continuous. (Keller and Saklas 1984) developed a procedure to estimate Passenger Car Equivalent
(PCE) values using simulation model. (Banks et al. 2000) mentioned the choice of whether to use a
discrete or continuous simulation model is a function of the characteristics of the system and the
objectives of the study. (Hossain 2001) used micro simulation technique to model traffic operations at
signalised intersections. (Arasan and Koshy 2005) developed a simulation model for urban roads.
(Mallikarjuna and Rao 2006) developed a Cellular Automata based simulation model to estimate PCU.
(Arasan and Krishnamurthy 2008) conducted a study on the effect of traffic volume and road width on
PCU values of vehicles using Microscopic simulation. (Arasan and Vedagiri 2006) applied the simulation
model to estimate the saturation flow rate of heterogeneous traffic. (Gowri and Sivanandan 2008)
developed a simulation model and examined the effects of left turn channelization on vehicle waiting
times. (Arasan and Arkatkar 2010) and (Arasan and Dhivya 2010) also have developed micro simulation
models for midblock sections. (Radhakrishnan and Mathew 2011) developed a traffic simulation model
integrating the concepts of cellular automata. Thus microscopic simulation models are used in recent
times for numerous purposes. However, the effect of signal time and signal phase modification can alter a
lot of changes to traffic properties of an intersection which has not been extensively studied.

TR31-2
Traffic signals are one of the most effective active controls of traffic. The advantage of traffic signal
includes an orderly movement of traffic and increased capacity of the intersection. However, the
disadvantages of the signalized intersection are large stopped delays and complexity in both design and
implementation. For a signalized section, signal phase is the part of a cycle allocated to a stream or a
combination of two or more streams having the right of way simultaneously. Traffic fixed time controller
signal may be of different phasing options. Figure 2 illustrates the conventional signal phase systems.
This study discuss various alternatives among signal phase and timing modification of fixed time
controller and their comparative effects on traffic condition such as congestion.

Figure 2: Traffic movements in two phase and four phase fixed-time controller signal system respectively
(Mathew 2017)

3 Methodology

The efficiency of a transportation network depends on the performance characteristics of its various
components. To improve the efficiency, proper planning and design of various facilities have to be
implemented (Sheela 2014). This paper focuses on simulating traffic movements at signalised
intersection by developing a micro simulation model using various signal phase system to improve the
intersection performance.

The studied intersection is modeled with micro simulation software named ‘VISSIM’ version 5.40. As
micro simulation encounters with individual track of vehicles, distance and time headway, lane changing
behaviour, car following model, so extensive data are required. Basic data collected are the geometry of
the intersection, number of lanes, lane width, median width, channel dimensions, traffic volume,
directional count, signal phasing, cycle times, green, red and amber times, saturation flow, traffic
composition, queue length, travel time, speed of vehicles etc. The traditionally simplest approach for
conducting manual count is to record each observed vehicle with a tick mark in a tally sheet that has been
used in this study. Indirect method of manual counting, the video camera is also utilized for the purpose of
verifying the collected data (Manual of Transportation Engineering Studies 2010) as presented in Figure
3.

Figure 3: Manual method (direct and indirect both) is used to collect traffic volume and intersection counts

TR31-3
With these data input, the intersection model is generated. The network generation process includes
importing and scaling overlay image after fixing units, determining driver behavior parameters, individual
link behavior fixing, link and connector generation, determination of vehicle types and classes, vehicle
composition selection, vehicle input in all the arrival points of the network, routing for various links, conflict
area and priority implementation and signal head along with signal program implementation etc. Then,
data collection points, travel time sections and queue counters establishments are done where necessary
for evaluation purpose.

After the network design completion, the simulation is initiated with random seed value. If the random
seed varies, the stochastic functions in VISSIM are assigned a different value sequence and the traffic
flow changes. This allows stochastic variations of vehicle arrivals in the network. For meaningful results it
is recommended to determine the arithmetic mean based on the results of multiple simulation runs with
different random seed settings.

Calibration of the generated model is executed to check whether it replicate the existing situation of the
intersection. Calibration is defined as the process of adjusting the parameters used in the model to
ensure that it accurately reflects input data. The subsequent process of validation is to run an
independent check on the calibrated model. Two sets of observed data are therefore required during the
model development process. One is used to calibrate the model by adjusting the parameters to ensure
that the output matches observed data and the second is used to verify that the aspects of the
performance of the calibrated model are in agreement to the set of observed data (Sykes 2010). It is very
significant that the output found for particular traffic parameters combinations are within the calibration
target range. Calibration target is shown in Table 1.

Table 1: Calibration target

Criteria and measures Acceptability targets

Hourly flows, model vs observed
Individual link flows
Within 15% for 700 vph < Flow < 2700vph >85% cases
Within 100 vph for Flow < 700 vph >85% cases
Within 400 vph for Flow > 2700 vph >85% cases
Total link flows
Within 5% All accepting links
GEH statistic- Individual link flows
GEH < 5 >85% cases
GEH statistic- Total link flows
GEH < 4 All accepting links
Travel times, model vs observed
Journey Times Network
Within 15% (or 1 minute, if higher) >85% cases
Visual audits To analyst’s satisfaction
Individual link speeds
Visually acceptable speed-flow relationship To analyst’s satisfaction
Bottlenecks
Visually acceptable queuing To analyst’s satisfaction

*Source: FHWA Freeway Model Calibration Criteria (Dion et al. 2012)

After the calibration and validation of the generated model is done, the signal phase and timing is
redesigned and modified. The existing model with prevailing two phase signal system is checked for
modified two phase and four phase signal system. For phase design, saturation flows at all the
approaches and lost times are determined first. Subsequently, the green period of all phases and cycle
time are calculated. Various traffic parameters are analyzed according to the changes. Alternatives are

TR31-4
compared and discussed whether the modification or measure will be suitable and efficient in terms of
performance increase of the Banglamotor intersection.

4 Data Collection

Banglamotor is a four legged intersection where Northbound, Southbound, Westbound and Eastbound
roads are labelled as SAARC Foara road, Shahbag road, Link road and New Iskaton road respectively.
East and Westbound roads have three lanes while South and Northbound roads having four lanes (Figure
4). It is to be mentioned that Eastbound road has a dedicated left turning lane. North and Southbound
roads carrying larger traffic flow hence considered as major road arterial.

Figure 4: Geometry of Link road (Westbound), New Iskaton road (Eastbound), Shahbag road
(Southbound) and SAARC Foara road (Northbound)

Typical data for the vehicle count for peak hour on a working day along with vehicle composition is shown
in Table 2 for the Shahbag road. Table 3 represents the Passenger Car Equivalent (PCE) values which
are used to calculate total volume in terms of PCE.

Table 2: Vehicle count for Shahbag road (Southbound)

Movement Motorcycle Bike Car/Jeep/ CNG Bus Utility Rickshaw/ Total %

Micro/Laguna Van Van
Left 18 5 41 15 0 0 0 79 3.72
Through 32 11 1074 374 428 2 0 1989 93.73
Right 2 1 25 14 10 2 0 54 2.54
Total = 55 19 1170 423 451 4 0 2122
% 2.59 0.90 55.14 19.93 21.25 0.19 0
PCE 41.25 9.5 1170 317.25 1353 4 0 2895

Total PCE 2895 passenger car per hour

Table 3: PCE value of vehicles (Partha 2009)

MoC6
Vehicle Categories
(2001)
Passenger car 1.0
Truck 3.0
Light good vehicle (Utility van) 1.0
Bus 3.0
Auto-rickshaw (CNG) 0.75
Motorcycle 0.75
Pedal cycle (Bike/rickshaw/van) 0.50

TR31-5
In the field it is observed that two phase fixed controller signal system prevailed but it was implemented
manually by traffic police. As a result, variation in cycle time often occurred. For the calibration purpose
the signal data taken as average of three cycle time observed in the field within the studied hour.

Total cycle time is found 241 sec, among which 190 sec of green time allocated for major road. As per the
calculation, the cycle time distribution for the considered phase systems is displayed in Table 4. The
generated model is illustrated in Figure 5.

Table 4: Cycle time distribution for existing, modified two phase and four phase system

Phase Cycle time Effective green Effective green Total Lost Time,
(sec) time (E-W) in sec time (N-S) in sec L in sec
Existing two phase 241 190 45 6
Modified two phase 200 162 34 4
Four phase 200 18(E) 18(W) 82(N) 72(S) 10

SAARC Foara Road (Northbound)

Link Road (Westbound)

New Iskaton Road
(Eastbound)

Shahbag Road (Southbound)

Figure 5: VISSIM 3-D and 2-D Model respectively used in the study

5 Model Calibration and Validation

To achieve high credibility for a simulation model, calibration and validation are of utmost importance
(Park 2005). The studied model conform the calibration targets as shown in Table 1, where the limits are
within the specified range. The heterogeneity of flow is eliminated by converting all traffic to PCE. The
pedestrians are not simulated in the study. Non-lane based flow stands in sharp contrast to micro
simulation and rigid in associating vehicles with specific lanes. So, driving behaviour parameters are
modified since traffic flow does not follow lanes unlike North American condition. The main parameters for
incorporating non-lane based flow are desired position at free flow, lateral distance, diamond shape
queuing, consider next turning direction and overtake on same lane. The other parameter’s sensitivity
analysis is performed. Finally using trial and error method, the calibrated parameters set is found that
mimics the field result in terms of flow and travel time. The parameter combination is displayed in Table 5.
Comparison of real data observed in the field and simulated data from VISSIM is illustrated in Table 6(a)
and (b). The calibrated model is validated for data collected from another working day.

6 Results and Discussion

After studying it is seen that at field condition there is considerable amount of average delay and queue
length generation. The average travel time of the vehicles in between the various links is moderately high
as well. Especially the right turn from Sahbag to New Iskaton road, face considerable amount of
congestion due to the conflict situation with the through movement of traffic coming from SAARC Foara
road. Link road and New Iskaton road traffic constitutes minor traffic flow so the signal timing modification
for major and minor road may be one solution to improve the present situation.

TR31-6
To achieve better performance in this intersection, various road traffic parameters, characteristics and
measures need to be experimented. In this study signal time modification has been implemented and
results are compared. Finally the best alternative in terms of efficiency and of the intersection is
suggested.

Table 5: Calibrated and modified parameters combination

Parameter Default Value Modified

Value
Significant Driving Behaviour Parameters:
Look ahead distance Min: 0 m, Max: 250 m 30
Look back distance Min: 0 m, Max: 150 m 30
Smooth close-up Behaviour unchecked checked
Average standstill distance 2m 2.45 m
Additive part of safety distance 2m 1.90 m
Multiple part of safety distance 3m 2.98 m
General behaviour Free lane selection as it is
Waiting time before diffusion 60 s as it is
Minimum headway (front/rear) 1m 0.5 m
Desired position at free flow Middle of lane any
Keep lateral distance to vehicle on next lane unchecked checked
Diamond shaped queuing unchecked checked
Consider next turning direction unchecked checked
Collision time gain 2.6 s 3.0 s
Overtake on same lane unchecked checked
Minimum lateral distance 1.0 0.90

Table 6 (a): Observed data in field v/s data from simulation

Direction Individual Diff. GEH statistic, Remarks

link flow |M-C|
∗
Field, Model,
C M
From SAARC Foara 3336 3254 82 1.43 Difference within 400 veh/hr, for
To SAARC Foara 3152 3062 90 1.61 Flow > 2700 veh/hr and within 100
From Shahbag 2895 2905 10 0.19 veh/hr, for Flow < 700 veh/hr for
To Shahbag 3668 3526 142 2.35 more than 85% of the cases.
From New Iskaton 504 526 22 0.97 GEH Statistic < 5 for Individual
To New Iskaton 261 224 37 2.33 Link Flows for more than 85% of
From Link road 543 535 8 0.34 the cases
To Link road 194 185 9 0.59
Sum 14553 14217 336 2.80 Sum of All Link Flows within 5% of
(2.31%) sum of all link counts and GEH < 4
for sum of all link counts.

Table 6 (b): Observed data in field v/s data from simulation

Direction Travel Times Difference Percentage Remarks

field model
From SAARC Foara 0.69 0.8 0.11 15.94 Road section considered: 10
To SAARC Foara 0.79 0.9 0.11 13.92 meter
From Shahbag 0.89 0.8 0.09 10.11
To Shahbagh 0.80 0.9 0.10 12.50 Journey Times, Network:
From New Iskaton 0.76 0.8 0.04 5.26 within 15% (or 1 min, if higher)

TR31-7
To New Iskaton 0.94 0.9 0.04 4.26 for more than 85% cases
From Link Road 1.67 1.9 0.23 13.77
To Link Road 0.94 0.8 0.14 14.89

With the existing two phase signal system, if average travel time is considered, it is observed from
simulation that modified two phase signal system serves the intersection well for almost all the cases.
Typical values are shown in Table 7. For instance, from Shahbag to New Iskaton, the travel time
decreases 369 to 318 sec. On the other hand, the four phase signal system generates more time than the
prevailing condition at present at a number of cases, making it an ineffective alternative. It is to be
mentioned that VISSIM evaluation result of travel time is the average travel time (including waiting or
dwell times) determined as the time required by vehicles between crossing the first and crossing the
second cross sections (VISSIM manual 5.40, PTV, 2012)

Table 7: Typical comparison of travel time for the generated model

Origin-Destination Distance Travel Time

in meter in sec
Existing two Modified two Four phase
phase signal phase signal signal
SAARC Foara to Shahbag 886.4 92.10 84.70 227.90
Shahbag to New Iskaton 1371 369.00 318.60 295.00
New Iskaton to SAARC Foara 1363 235.00 220.50 231.40
Link road to Shahbag 622.1 146.30 134.50 322.40

* Travel time refers to the average travel time (including waiting or dwell times)

From simulation, average total delay per vehicle, average standstill time per vehicle and average number
of stops per vehicle is calculated. The total delay is computed for every vehicle completing the travel time
section by subtracting the theoretical (ideal) travel time from the real travel time. The theoretical travel
time is the time that would be reached if there were no other vehicles and no signal controls or other
stops in the network. Comparison of delay statistics for the alternative solutions studied are typically
illustrated in Table 8. The modified two phase system appears to be a feasible improvement in terms of
delay minimization.

Table 8: Typical Comparison of delay for the generated model

Direction Existing two phase Modified two phase signal Four phase
signal signal
Delay Stopd Stops Delay Stopd Stops Delay Stopd Stops
in sec in sec
SAARC Foara to 9.20 5.50 0.36 6.50 3.50 0.22 153.60 134.90 1.96
New Iskaton
From Shahbag to 252.10 192.20 9.44 201.20 152.70 8.28 177.50 161.90 1.79
New Iskaton
Link road to New 102.30 92.20 1.63 104.20 94.30 1.53 367.50 341.70 3.58
Iskaton
New Iskaton to 118.90 102.70 1.97 104.40 89.00 2.03 115.00 105.20 1.27
SAARC Foara

*Delay is the average total delay per vehicle, Stopd is the average standstill time per vehicle and Stops is
the average number of stops per vehicle

In VISSIM, queues are counted from the location of the queue counter on the link or connector upstream
to the final vehicle that is in queue condition. For Link from SAARC Foara, queue length improves a lot as
shown in Table 9 for modified two phase system. Although maximum queue length at link from Shahbag

TR31-8
increases from to 337 to 391 m. The minor roads do not show significant improvement in terms of queue
length generation. Four phase system yet again fails to satisfy improvement, as it creates longer queue
for almost all approaches.

Table 9: Typical Comparison of queue length for the generated model

Direction Existing two phase Modified two phase Four phase

signal signal signal
Queue Max queue Queue Max queue Queue Max queue
length length length length length length
From SAARC Foara 37 264 17 115 345 505
From Shahbag 85 337 85 391 331 505
From New Iskaton 25 72 23 67 25 66
From Link road 43 115 38 116 182 472

* Queue length is the average queue length and queue length is output in units of length not in number of
cars

Finally the network performance of the model is checked. The average delay time per vehicle, average
speed, total delay and travel time are simulated as displayed in Table 10.

Table 10: Typical Comparison of network performance for the generated model

Term Existing two phase Modified two phase Four phase

signal signal signal
Average delay time per vehicle in sec 34.75 29.49 240.36
Average speed in km/hr 28.81 30.25 14.56
Total delay time in hr 69.69 59.14 479.66
Total travel time in hr 224.55 213.95 737.30

The comparison of various data extracted by simulation among the three alternatives (existing two phase,
modified two phase and four phase signal system) discussed, it is obvious that four phase system will not
be suitable at all and it will worsen the present condition. Most of the parameters suggest that modified
two phase system will be more effective in elevating the performance of the studied intersection.

7 Conclusions

Signalized intersections are key elements in the urban transportation network. Increasing traffic demand
can exceed the carrying capacity of the intersection during peak periods. Consequently, traffic condition
deteriorates. The studied intersection has started facing congestion problem and measures should be
taken to improve the prevailing condition. The complex traffic situation prevailed in this study is non-lane
based and heterogeneous. The real field condition is simulated by modifying the default driving behavior
characteristics of VISSIM. This modification is significant to represent the driving behavior that does not
respect lane discipline. Also all the vehicles are converted to equivalent passenger car to induce
homogeneity. As per study, it is evident that the modification of existing two phase signal system can be
a great alternative to facilitate better performance to the intersection. Four phase signal system
implementation will not be suitable as far as the study results suggest. Existing two phase signal’s cycle
time modification eliminates congestion problem to a great extent and flourish other parameters such as
lesser travel time and queue length etc. The model can be tested for other traffic parameter changes
other than signal time modification. However, it is obvious that signal time modification can be a feasible
alternative to improve performance of the studied intersection. This technique to improve intersection
proficiency can be implemented to any North American intersection. Although the driving behavior

TR31-9
parameters need to be adjusted due to the fact that there generally exists lane based homogenous or
heterogeneous flow. Congestion in intersections is a waste of money and resources. The studied
strategy can be applied to fabricate efficacious intersection that would lead to an efficient urban traffic
system.

Acknowledgements

Thanks to Almighty Allah for His graciousness. The author expresses gratitude to Prof Dr. Md.
Hadiuzzaman, Prof Dr. Hasib Mohammad Ahsan and Anup Paul for their continuous assistances. Thanks
to Head of the Civil Engineering department and Commandant of MIST for the encouragement. A very
special gratitude is offered to the author’s parents, wife and his younger sister for their support.

References

Arasan, V.T. and Arkatkar, S.S. 2010. Microsimulation study of effect of volume and road width on PCU
of vehicles under heterogeneous traffic. Journal of Transportation Engineering, ASCE, 136(12): 1110-
1119.
Arasan, V.T. and Dhivya, G. 2010. Methodology for Determination of Concentration of Heterogeneous
traffic. Journal of Transportation Systems Engineering and Information Technology, 10(4): 50-61.
Arasan, V.T. and Koshy, R.Z. 2005. Methodology for Modeling Highly Heterogeneous Traffic Flow.
Journal of Transportation Engineering, ASCE, 131(7): 544-551.
Arasan, V.T. and Krishnamurthy, K. 2008. Study of the Effect of Traffic Volume and Road Width on PCU
Values of Vehicles Using Microscopic Simulation, Journal of the Indian Roads Congress, 69(2): 133-
149.
Arasan, V.T. and Vedagiri, P. 2006. Estimation of saturation flow of heterogeneous traffic using computer
simulation. Proceedings of the 20th European Conference on Modelling and Simulation.
Banks, J.; Carson II, J.S.; Nelson, B.L.; Nicol, D.M. 2000. Discrete-Event System Simulation. Prentice.
Hall of India Private Ltd., New Delhi.
Dion, F., Karthik, S. and Ban, X. 2012. Evaluation of Traffic Simulation Model Use in the Development of
Corridor System Management Plans (CSMPs). California PATH Research Report, UCB-ITS-PRR-
2012-2.
Gowri, A. and Sivanandan, R. 2008. Evaluation of left turn channelization at a signalized intersection
under heterogeneous traffic conditions. Transport, 23(3): 221-229.
Institution of Transportation Engineers. 2010. Manual of Transportation Engineering Studies, ITE, 2nd
edition, Washington, DC 20006 USA.
Keller, E.L. and Saklas, J.G. 1984. Passenger Car Equivalents from Network Simulation. Journal of
Transportation Engineering, ASCE, 110(4): 397- 411.
Mallikarjuna, Ch. and Rao, K.R. 2006. Area Occupancy Characteristics of Heterogeneous traffic,
Transportmetrica, 2(3): 223-236.
Mathew, V. 2017. Design Principles of Traffic Signal. Lecture Notes in Transportation Systems
Engineering, IIT Bombay (tvm@civil.iitb.ac.in).
Naghawi, H. and Idewu, W. 2014. Analysing delay and queue length using microscopic simulation for the
unconventional intersection design Superstreet. Journal of the South African Institution of Civil
Engineering, 56(1), 100–107.
Partha, S., Hossain, Q., Iqbal, H. and Islam, M. Passenger Car Equivalent (Pce) Of Through Vehicles At
Signalized Intersections In Dhaka Metropolitan City, Bangladesh. IATSS RESEARCH, 33-2
Park, B. and Qi, H. 2005. Development and Evaluation of a Procedure for the Calibration of Simulation
Models. Transportation Research Record: Journal of the Transportation Research Board, No. 1934,
Transportation Research Board of the National Academies, Washington, D.C., pp. 208–217.
PTV, V. 2012. VISSIM 5.40 - User Manual, PTV, Karlsruhe, Germany.
Radhakrishnan, P. and Mathew, T.V. 2011. Passenger car units and saturation flow models for highly
heterogeneous traffic at urban signalised intersections. Transportmetrica, 7(2): 141-162.
Sheela, A. and Kuncheria I. 2014. Traffic Simulation Model And Its Application For Estimating Saturation
Flow At Signalised Intersection. International Journal for Traffic and Transport Engineering, 4(3): 320 –
338.

TR31-10
Sunil K. and Ranjitha, J. 2013. Improvement of Traffic Operations in Congested Signalized Intersections -
A Case Study In Bangalore City. International Journal of Engineering Research & Technology (IJERT),
2(7), ISSN: 2278-0181.
Sykes, P. 2010. Traffic Simulation with Paramics. In J. Barcelo, ed. Fundamentals of Traffic Simulation.
Barcelona, Springer. pp.131-72.

TR31-11