DEVELOPMENT OF TRAFFIC FLOW MODEL ON SUBURBAN ROAD

DONE BY- RUPAYAN MONDAL

(ENROLLMENT ID.-510415041)
UNDER GUIDENCE OF – Prof. PRITAM SAHA
DEPARTMENT OF CIVIL ENGINEERING
IIEST, SHIBPUR
HOWRAH, 711103
INTRODUCTION

 In our daily life, there are numerous essential services that we can’t
overlook, roadway service is one of them.
 It is necessary to study the various characteristics of this roadway traffic
to organize and operate the service properly and efficiently. But it is not
so easy due to its complex nature.
 So, we need some computerized exploration called SIMULATION to
comprehend the traffic nature in a handsome way.
 Interest in modeling traffic flow has been around since the appearance
of traffic jams. Ideally, if we can correctly predict the behavior of vehicle
flow given an initial set of data, then, in theory adjusting the flow in
crucial areas can maximize the overall movement of traffic along a
stretch of road.
SIMULATION

 Simulations models are the representation of the real world systems in a

mathematical and logical way.
 Traffic simulation is analysis of the traffic situation in a systematic,
computerized and experimental way through some traffic models, which
gives us the concept of traffic propagation.
 Traffic simulation can be classified as per the interaction between driver-
driver or driver-road-
1. Microscopic
2. Macroscopic
3. Mesoscopic
 Microscopic simulation deals with individual vehicle and their interaction
at a higher level of details.
 In microscopic simulation traffic flow is captured by representation of
travel demand and behavior(e.g. route choice models) and driving
behavior(e.g. speed/acceleration, lane changing and gap acceptance
models).
 The realism of the simulator depends on the richness and reliability of
these models and therefore development of sound behavior models and
rigorous calibration and validation of these models are critically
important to any application of the simulation tool.
 Microsimulation is an inevitable tool for analysis and interpretation of
real world situation.
 Some well known simulation models developed by the researchers are-
SOVT model( WU & Heimbach,1981),
SIMTOL model( Stock & May,1976),
TRARR model(Robinson,1980),
CARSIM model(Benekohal & Treiterer,1988) etc.
VEHICLE TRAJECTORY

 A vehicle trajectory is defined as the path taken by a vehicle over a

period of time and is determined by projecting its known lateral position
in the lane from the pre determined stations.
 Modeling of vehicular travel behaviors with interactions also the
calibration and validation of existing model such as car following and
lane changing behaviors, counts heavily on multiple vehicle trajectory
data.
 Other important parameters severely contribute to microscopic
simulation modeling, such as velocity, acceleration and headway, can
be derived from vehicle trajectory data.
BACKGROUND

 In developed countries, the prevailing traffic is predominantly composed

of passenger cars and is generally called as ‘homogeneous’ traffic.
However, in most of the less developed countries, traffic is found to be
composed of a wide variety of vehicle types including two and three-
wheeled vehicles and also, non-motorized vehicles.
 Since, the static and dynamic characteristics of these vehicles are
considerably different; they share the same road space without any
physical segregation and eventually, result in a traffic stream which is
fundamentally different than that of ‘homogeneous’ traffic, known as
‘heterogeneous’ traffic
 Most of the national highway and state highway system in India is of two lane
and traffic flow on Indian highway is quite interesting to be studied because of
two reasons.
 First, the multi-class traffic with vehicles such as cars and pickups having high
maneuverability and heavy vehicles such as trucks and buses having low
maneuverability. The speed of these vehicles vary from 40 to 100 km/h.
 Second, quasi-lane disciplined traffic movement on Indian road, with some
vehicles following lane discipline and many others not. Moreover, its
composition and the level of lane discipline may change significantly based on
the traffic flow-level.
 When such different types of vehicles with varying static and dynamic
characteristics are allowed to mix and move on same roadway facility, a variable
set of longitudinal and transverse distribution of vehicles may be detected from
time to time.
 The behavior of the mixed traffic being very interactive, neither a mathematical
nor a physical model is amenable for carrying out experimental studies.
Computer simulation modeling is better suited for studying various traffic
characteristics, as they are more efficient, economic and flexible.
SCOPE OF THE STUDY

 The better understanding of traffic flow can be used to further

characterization of heterogeneous traffic, to evaluate alternative
treatment, as an element of design procedure and also for safety
analysis.
 Now a days industrialized nations are looking forward promoting non-
motorized slow vehicles as environment-friendly modes of transport.
Application of these vehicles may demand heterogeneous traffic model
to other context.
 Hence, further development of traffic flow model will help the transport
professionals to design effective and appropriate control systems and
better estimates the vehicular emissions for heterogeneous traffic.
OBJECTIVES OF THE STUDY

As mentioned above studying the heterogeneous traffic over suburban

area is highly challenging and difficult. An appropriate simulation technique
can be used to characterize such difficult and unpredictable traffic flow.
Accordingly, this study aims at meeting the following objectives-
1. Studying lane changing, crossing and overtaking behavior on suburban roads
using vehicle trajectories over space and time
2. Developing a microsimulation model from the prospective of observed
vehicle trajectory and investigating the impact of petrol/diesel/LPG/battery-
run vehicular mix on operation and safety of traffic
LITERATURE REVIEW

Simulation has been acknowledged as one of the best tools for modeling
the traffic flow under both homogeneous and heterogeneous conditions.
Here are some literature reviews on traffic simulation-
 MARWA and RAMASESHAN (1978): They used the simulation
technique to study the vehicular interaction in heterogeneous i.e. in
mixed traffic flow.
The simulation was framed for two lane two way traffic which was
included with overtaking and restricted stream logics.
Speed volume relationship was developed for vehicle
combinations and LOS was calculated in terms of operating
 RAMANAYYA (1988): He developed simulation model for mixed traffic flow on
single-lane one way, single-lane two way and two-lane two way roads.
The model assumed car following logic to maintain safe gap and also assumed
that any fast vehicle would overtake the slower one, provided the lateral clearance is
higher than prescribed minimum value.
Validation of the model was based on comparison observed and simulated
speed distribution.
 GUNDALIYA et al (2010): The model adopted here, used the concept of
minimal modelling aspect of cellular automata. This aspects were modified for
heterogeneous traffic flow dividing road stretch in grid of cells.
In this model, various traffic characteristics like acceleration, deceleration and
maximum speed of vehicles were incorporated. The position and speed of the
vehicles are assumed to be discrete and the speed changes according to its
interactions with other vehicles.
The model was calibrated and validated using real data and VISSIM.
The results indicates that grid based model can represent the complex
heterogeneous traffic almost properly. But the model requires more testing at
different traffic conditions.
 PPDEY et al. (2008): A computer model was developed by them to simulate the
traffic flow from several location of two lane roads in India incorporating all traffic
characteristics like speed, placement, arrival, acceleration, and overtaking
characteristics of different types of vehicles.
After determining the capacity and studying the effect of traffic mix over
capacity and speed they concluded capacity reduces as the directional split moves
away from the even one and also as proportion of the three wheeler, tractor, or
heavy slow vehicles increases in traffic stream.
The model can provide capacity and equivalency factor for different vehicles on
different level of traffic mix and lane directional distribution but it needs further
development to take into account the effect of shoulder condition and road width.
 S.ARKATKAR et al (2016): They aimed to implement traffic flow model on
Indian urban expressway specifically Delhi-Gurgaon expressway and to evaluate
its capacity using micro-simulation model VISSIM 5.40.
The model was used to develop fundamental traffic-flow relationship, namely,
speed-flow, speed-area occupancy and flow-area occupancy for the traffic flow levels
starting from nearly-zero to its capacity.
The simulation study suggested that for characterizing the effect in change in
traffic composition on traffic flow characteristics for an expressway, the speed flow
relationship is the most appropriate.
The study further showed that in mixed traffic condition, the capacity-level
critical speed reduces linearly with increase in percentage of trucks from 10 to 100%
in the traffic stream, comprising only two vehicle categories: cars and trucks.
The driver’s behavior considered in this study can be refined further to consider
many more physiological and psychological factors.
 HENG WEI et al (2005): The paper presents a methodology for
extracting vehicular trajectory data using a video capturing technique
and also to label a computer based tool, VEVID (vehicle video capture
data collector), which was developed to extract the vehicular trajectory
data from the video.
There are several steps mentioned to achieve the proposed
objectives. For microscopic modeling vehicular trajectory data is highly
significant and the methodology proposed here is rather economic and
efficient.
The limitation of the proposed method lies in the extracting
trajectory data of the vehicle located over 400m from the observation site.
 V ARASAN et al (2010): T adopted a dynamic stochastic type discrete event
simulation in which the aspects of interest over heterogeneous traffic flow are
analyzed numerically with the aid of computer program.
The model was applied to measure one of the fundamental characteristics of
traffic flow i.e. concentration. It is a border term including both density and
occupancy.
In this paper they proposed a new concept named ‘area-occupancy’ to
measure traffic concentration.
It was found by using and validating simulation model, that the new concept
area-occupancy is a valid measure and can represent the traffic concentration under
heterogeneous condition.
MOTIVATION

 It is clearly noticeable from the literature reviewed in the previous

section that although reasonable number of experimental studies and
modeling have been done on homogeneous traffic flow and some sort of
heterogeneity over urban roadway.
 But till now there is no such model available which can characterize the
most complex heterogeneous traffic flow over suburban road where
huge fluctuation over static and dynamic characteristics of the vehicles
occurs.
 This motivated the experimental study of this study.
PRPPOSED METHODOLOGY

The usual roadway and traffic flow conditions on suburban road in India are enormously different when compared with other roadway
in India, additionally there is no perfect lane-discipline at all. The knowledge over traffic condition is no doubt a vital input for planning,
designing, analysis and operation of roadway system. So, here is some methodology proposed aiming to model traffic flow over suburban
road.
 Firstly, we have to identify and acknowledge the basic input parameter required for the simulation model. These should be provided
during the simulation run otherwise it will become hectic. After identifying the basic parameters we have to measure that parameters
from some field or secondary data.
 On the basis of the selected parameters, we have to determine types of data required. To collect the data first we need to finalize the
study road stretch after conducting a reconnaissance survey. We have to collect data from the field using video recording method for
several times and extract the input data from that video using some data extracting software. Now the extracted data would be analyzed
to develop multiple vehicle trajectory data. These vehicle trajectory data are required to develop the microsimulation model.
 Then we have to prepare the microsimulation model using some software. Model development involves three steps-(1) defining model
parameters, (2) calibration of the simulation model, (3) validation of the simulation model.
 In the first step in model development different model parameters are defined like speed distribution, traffic composition and flown etc.
Calibration is a process of refining the model to replicate observed data and observed site conditions to a sufficient level of accuracy in
order to satisfy the model objectives. This process involves refining the characteristics, namely desired speed distribution,
acceleration/deceleration of vehicle, mechanical characteristics of the vehicle, minimum safety distance, minimum lateral distance, and
driving behavior characteristics. By giving these parameters as an input to simulation model, simulation runs have to be carried out in
order to estimate the output. Validation of the model can be defined as the process of checking the results obtained from the calibrated
model in terms of simulated values against field measurements for parameters.
 After application of the model and estimating the output we have to investigate the impact of petrol/diesel/LPG/battery-run vehicular
mix on operation and safety of traffic.
 FLOW CHART OF PROPOSED METHODOLOGY

Investigate the
Identify the Preparation of
Development of vehicular mix
a
input input
parameters
microsimulation Estimate output impact over
parameters model operation and
safety of traffic

Data extraction Model

Data collection Model validation
and analysis calibration
REFERENCES
 Khan, S. I., and Maini, P. (2000) “Modeling heterogeneous traffic flow.” Transportation Research Record, 1678, Transportation
Research Board, Washington, D.C., 234–241.
 Ramanayya, T. V. (1988) “Highway capacity under mixed traffic conditions.” Traffic Eng. Control, 29, 284–87.
 Dey, P. P., Chandra, S. and Gangopadhyay, S. 2008. Simulation of mixed traffic flow on two-lane roads, Journal of Transportation
Engineering ASCE, 134, 361–369.
 Arasan, V. T. and Koshy, R. Z. 2005. Methodology for modeling highly heterogeneous traffic flow, Journal of Transportation
Engineering, 131, 544–551.
 Gundaliya, P. J., Mathew, T. V. and Dhingra, S. L. 2010. Heterogeneous traffic flow modelling for an arterial using grid based
approach, Journal of Advanced Transportation, 42, (4), 461–491.
 Marwah, B. R., and Ramaseshan, S. (1978). “Interaction between vehicles in mixed traffic flow using simulation.” Highway Research
Bulletin No. 8, Indian Roads Congress, New Delhi, 1–16.
 Arasan, V.T. and Dhivya, G. (2010). “Simulation of highly heterogeneous traffic flow characteristics”
 Arkatkar, S.S., Velmurugan, S., Puvvala, R. and Narula, S. (2016) “Methodology for simulating heterogeneous traffic on expressway
in developing countries: a case study in India”
 “Fundamental of traffic simulation” by Jaume Barcelo
 Mukherjee, S. K., Rao, S. K., and Raichowdhury, M. L. ~1988!. “Fitting a statistical distribution for headways of approach flows at two
street intersections in Calcutta.” J. Inst. Eng. (India), Part AG, 69, 43–48.
 Marwah, B. R., and Singh, B. ~2000!. “Level of service classification for urban heterogeneous traffic: A case study of Kanpur
metropolis.” Transp. Res. Circular E-C018: Proc., 4th International Symp. On Highway Capacity, Maui, Hawaii, 271–286.
 Nagraj, B. N., George, K. J., and John, P. K. ~1990!. “A study of linear and lateral placement of vehicles in mixed traffic environment
through video-recording.” Highway Res. Bulletin, No. 42, IndianRoads Congress, New Delhi, India, 105–136.
 
THANK YOU