PSH CLT301 Undergraduate
Delivering Tomorrow's
Solutions Today System Development Project
Penang Bridge Dynamic Traffic
Simulation Tool
System Document
(SPMP & SRS)
For:
Mr Wan Mohd Nazmee Wan Zainon
By:
NOVA
Muhammad Ammar Firdaus bin Md Anuar - 95959 ( Team and Project Leader )
Ibnu Bukharie bin Ismail - 95906 ( Quality Assurance )
PPSK Software House
School of Computer Sciences
Universiti Sains Malaysia
11800 USM
Penang, Malaysia
Session 2009/2010
TABLE OF CONTENT
1. Introduction ...................................................................................2
2. Problem Statement ........................................................................2
3. Proposed Solution ..........................................................................3
4. Objective ........................................................................................6
5. Scope and system limitation ...........................................................6
6. Benefit of system ...........................................................................6
7. Literature Review ...........................................................................7
8. Software and Hardware Requirement .......................................... 10
9. Architectural Diagram .................................................................. 11
10. Module Diagram .......................................................................... 12
11. Task Division ................................................................................ 13
12. Methodology................................................................................ 14
13. Conclusion .................................................................................... 15
14. References ................................................................................... 16
1
1. Introduction
Since the year 1985, the island, Pulau Pinang has been connected to the mainland, Prai
by the famous Penang Bridge whereby it is one of the longest bridges in the world. Penang's
rapid economic boom especially on the island, has resulted in over than 120,000 vehicles using
the bridge daily. Moreover, the figure increases up to 10,000 more during the school holidays
period [1]. Every day, hundreds of thousands of Penang Bridge user are facing traffic congestion
as they commute to work in their vehicles through the Penang Bridge. The impact from this
traffic congestion is considered to be large in the terms of travelling time, resources, and
pollution. The vehicle trips per day are estimated to be 120,000 and it is also estimated that the
Penang Bridge congestion alone cost the travelers 328 million hours of delay, 10 million liters of
wasted fuel, and minimum RM140 million in toll fees annually [2]. A lot of measures have been
taken to solve the traffic problem in and out of Pulau Pinang including the widening project of
Penang Bridge.
Through the research, a dynamic traffic simulation is identified as an alternate solution
for reducing traffic congestion whereby it is capable of simulating the traffic flow first before
implementing any suitable measure. A computer simulation or computer model is a computer
program that attempts to simulate an abstract model of a particular system. On the other hand,
the term dynamic in the phrases dynamic traffic simulation tools is referring to the ability to
adjust the parameter values such as speed limit and number of vehicle while the simulation is
still running.
2. Problem Statement
Penang bridge Toll Plaza has delayed the traffic flow causing the traffic queue
Drivers coming from Lebuhraya Utara Selatan increased the length of traffic queue in
front of the Toll Plaza
If the speed of vehicles from both direction are too fast, they will arrived at the traffic
queue faster and have to stop
Therefore, the drivers have to wait for the earlier queue to clear first before they can
move their vehicle once again
This problem shouldn't happen if the drivers obeys a specific speed limit when they
approaching the congested area
2
3. Proposed Solution
Therefore, we would like to propose a dynamic simulation tool as a solution to reduce the
problem of traffic congestion that occurs at a given area of a merging point. We suggest that
this dynamic simulation tool is used as an approach to forecast or predict the traffic activity
whereby it is focused on the traffic flow at the merging point of the toll booth area. It also
comes with a suggestion tool that can suggest a suitable speed limit for drivers.
This is how the simulation area would look like in the application whereby it is arranged in a
more systematic form. For example, assume that the average speed of the vehicles on the
Penang Bridge is 30 km/hour which are considered as heavy traffic congestion. Due to the
traffic congestion at the bridge, the lane area of before Penang Bridge especially the area
before the Penang Bridge Toll Booth will become congested as well. If all the drivers would
simply speed up their vehicle at their own convenience, the traffic congestion will continue to
occur. Thus, the traffic queue would become much longer and more time is consumed before
the traffic flow become stable again. We have to bear in mind that most of the drivers don’t like
to wait still in a queue especially when it is a rush hour and the weather is unpleasant.
However, this can be prevented if the traffic flow is controlled from the very beginning. By
encouraging all the drivers that are planning to use the Penang Bridge an appropriate speed
limit, we can allow the earlier traffic congestion at the front to stabilize first before allowing the
3
next vehicles to move. This way, the traffic flow will become much smoother without any need
to wait in long queue.
The suggestion tool would calculate a suitable speed limit for all the drivers in the range of 10
km before approaching the bottleneck area. Then, it will show the speed limit result at the
electronic sign board so that the driver would be easily alerted by the message shown. This
should be a more effective method of enforcing the speed limit because the speed limit is
flexible whereby it can be changed based on the traffic condition of the queue at Penang Bridge
Toll Booth. Therefore, from time to time, the drivers are not completely restricted to follow the
same speed limit.
In addition, there are 2 algorithm are going to be use as an approach to produce a more
accurate simulation especially when handling the effect of merging traffic flow. They are known
as the fluid-dynamic algorithm and the kinematic wave theory. These 2 algorithms are chosen
for this simulation for a specific reason.
Algorithm Relevant Function
Fluid-dynamic algorithm Optimized the traffic flow on the road
Kinematic wave theory Optimized the lane-changing effect of vehicles on the road
Therefore, these 2 algorithms will be combined and integrated in the dynamic traffic to obtain a
better result of traffic activity such as travel time and speed limit in order to reduce the traffic
congestion at the given specific point.
4. Objectives
a) To implement the usage of fluid-dynamic algorithm and kinematic wave theory for a
more accurate calculation results.
b) To forecast the traffic congestion through a graphical computer simulation
c) To suggest an appropriate speed limit for the incoming drivers that plan to use the
Penang Bridge.
4
5. Scope and system limitation
The scope of this simulation system is specified as below:
Location Merging point of Penang Bridge Toll Booth
Type of vehicle Light vehicle and heavy vehicle
Traffic Parameter Value Velocity, speed limit, number of vehicles
However, this system has several limitations based on their scope whereby:
Limitation Description
Simulation area Simulation does not include the whole Penang bridge whereby the
area of simulation only took place at the merging point of Penang
Bridge Toll Booth.
Type of vehicle The simulation only limited to 2 type of vehicles: cars and trucks
Each represent light vehicle and heavy vehicle
Vehicles such as buses and motorcycles are not included
Vehicle speed Different vehicles move at different speeds on the road, so:
We assume that all car move at the same speed
Same assumption is made for trucks
6. Benefit of System
Short Term benefit
Reduce the traffic congestion to a minimum level
Reduce the travel time for Penang Bridge users
Allow any road user that plan to use Penang Bridge to be more alert
Long Term benefit
Promote healthy environment by reducing the air pollution and fuel consumption
Scenario planning idea can be tested first before it is executed which lead to a better
road construction in the future
Avoid the drop in the efficiency of workers to commute
Encourage a safety driving behavior among the drivers
5
7. Literature Review
In this chapter, we have made some research on the issues related to simulation of traffic
congestion which includes the causes and method of reducing the traffic congestion, the
simulation approaches and the algorithms used in the traffic simulation.
7.1. Traffic Congestion
Generally, traffic congestion occurs when the traffic demand is greater than the capacity of the
road. Traffic congestion is considered to be at a critical level if the vehicles are fully
incapacitated for a long period of time. In a report by the Federal Highway Administration
(FHWA) [3], traffic congestion can be also described as a stop-and-go traffic.
FHWA [3], have classified 6 main causes of traffic congestion whereby they often interact with
one another. These are the causes that been identified:
Causes Definitions
Physical Bottleneck / Capacity Capacity is the maximum amount of traffic that is capable of
being handled by a given highway section. Capacity is
determined by several factors:
a) The number and width of lanes
b) Grades and roadway alignment at merge areas
Traffic Accidents Event that disrupts the normal flow of traffic usually by
physical interference in the travel lanes. Event such as
vehicular crashes, breakdowns and debris in travel lanes are
a few common example of this incidents.
Work Construction Zone Construction activities on the roadway that result in physical
changes to the highway environment. These changes may
include a reduction in the number or the width of travel lane
and a temporary roadway closure.
Weather Natural environmental conditions can lead to changes in
drivers’ behavior in driving.
Traffic control device Intermittent disruption of traffic flow by control devices such
as railroad grade crossing and poorly timed signal also lead to
congestion and ravel time variability.
Special events A special case of demand fluctuations whereby traffic flow in
the vicinity of the events will be radically different from
typical patterns.
Causes of traffic congestion
6
Several approaches have been taken to reduce traffic congestion problems and one of them is
the Intelligent Traffic Management System (ITS). The table below shows the list of approaches
taken as countermeasure the traffic congestion problem.
Approaches Explanation
Intelligent Transportation Systems Includes Vehicle Monitoring System (VMS), Closed Circuit
(ITS) Television (CCTV), Traffic Operation Centers, Remote
Traffic Signal Control.
Road Pricing or tolls One of the examples would be is the London congestion
charge whereby it is a free levied on vehicle drivers
entering the centre of the city.
Cheaper and better public Encourage people to use public transport instead of using
transportation. their own vehicle.
Bus-only lanes Dedicated bus rapid transit roads.
Broadcasting of road condition Inform the road user via media such as cell phone.
Visual barrier “Rubbernecking syndrome” whereby curious driver are
slowing down their vehicles to observe events such as
accident on opposite lanes
Park and ride Encourages people to park outside the main town or city
and switch to using public transport.
Proper city planning Practices that avoid concentration of traffic on arterial
roads. One of the measures taken includes arranging
streets in fused grid geometry rather than a complex
branching tree-like network topology.
List of approaches taken to reduce traffic congestion
7.2. The Algorithms in Traffic Simulation
Many algorithms have been implemented in traffic simulation modeling. The type of algorithms
needed depends on the type of traffic environment evaluated. A certain efficient algorithms are
needed in order to solve the problem of traffic congestion regarding the merging bottleneck
during traffic flow. These 2 algorithms are the fluid-dynamic algorithm and the kinematic wave
theory. These algorithms are chosen because the fluid-dynamic algorithm can manage the
traffic flow pretty well whereby the kinematic wave theory optimizes the lane-changing effects
on vehicles.
7
7.2.1. Fluid- dynamic algorithm
In a study conducted by R. Natalini et al. [4], the authors discussed a simulation algorithm based
on a fluid-dynamic model. In the paper, numerical approximations were considered which
includes a brief discussion of boundary conditions and conditions at the junctions.
On another paper by G. Bretti [5], the author describe that there are 2 ways to measure
velocity. The 2 ways are:
Recording the velocity as whereby
= velocity, = distance, = time
Associate each point in space at each time a velocity field, whereby
= position and = time
Therefore, the basic fluid-dynamic algorithm can be represent by,
Traffic flow = traffic density x mean velocity [5]
Traffic flow = Average number of car passing per time unit
Traffic density = Number of cars per distance unit
If the velocity reach the maximum value, which means that there no other cars on highway and
this vehicle are able to move at highest speed. On the other hand, if the velocity reaches values
of 0, it means that the vehicle is not moving or also known as bumper to bumper traffic.
7.2.2. Kinematic Wave Theory
Carlos F. Daganzo [6], have studied a theory regarding a kinematic wave theory which is based
on a kinematic wave model of Lighthill and Whitnam (1995) and Richards (1956). It is stated
that this theory of traffic dynamics recognize that vehicles would took up space whereby it is
focused on queued traffic and mixed traffic.
In a paper by Lighthill and Whitnam [7], the authors studied on the kinematic wave theory
related to the flood movement. The wave property with regards to the flow and concentration
are described as kinematics. This paper uses the technique of kinematic waves to analyze the
flood movement in long rivers which could further be related to traffic system.
8
We Long Jin [8], proposed a kinematic wave theory of lane-changing vehicular traffic where
different lanes share the same traffic condition. The frequency of the lane changes among
vehicles in highway merging, diverging and weaving require a great study because bottlenecks
and accident frequently such area. In his paper, e modeled a lane changing traffic dynamic
based on the kinematic wave theories as follows:
[5]
p = density / number of vehicles per unit distance
e = lane changing effect parameter
v = velocity
This applies when traffic conditions and lane changing effect are almost constant. However, the
lane changing effect parameter is only functioning when they are difference traffic queue
length between 2 certain lanes.
Comparison between Fluid-dynamic algorithm and Kinematic Wave Theory
Algorithm Advantages Disadvantages
Kinematic Wave A process of reducing traffic queue Explicit simulation could generate a
Theory small fluctuation in travel time
Can be applied to a more Additional time steps are required if
complicated problem such as frequently boundary conditions are
traffic flow concerned
Fluid-dynamic A process of maximizing vehicle These model lacks the ability to
Algorithm flow backward propagation of shocks
Do not allow heavily congested
traffic flow
9
8. SOFTWARE AND HARDWARE PLATFORM
For the software part, we are using Netbeans IDE 6.5 as our platform to implement this project.
We are agreed to select this platform because it has a readymade windows form for developing
graphical user interface (GUI) that can be easily used. In addition, we are using SIM JAVA and
SILK executable file (.exe) as a simulation toolkit for simulation development. Moreover, it is
implemented on Windows XP with service pack 2. For the hardware part, we are running this
project at a computational speed of Core 2 Duo 1.8GHz and a memory of 2GB of RAM installed.
Below is the summarization of the hardware and software platform:
SOFTWARE PLATFORM
• Windows XP - service pack 2
• Netbeans IDE 6.5
• Simulation tool - Simjava, Silk, Javasim, JSIM, DEVS-JAVA, DESMO-J
HARDWARE PLATFORM
• Computational speed of Core 2 Duo 1.8GHz
• 2 GB Random Access Memory (RAM)
Figure 1: Software and Hardware Platform
10
9. ARCHITECTURE DIAGRAM
Figure 2: Architecture Diagram
10. MODULE DIAGRAM
Figure 3: Module Diagram of Traffic Simulation Tool
11
Based from our analysis and discussion on the traffic simulation tool, there are 2 sub modules
that will be our focus in implementing this project. There are simulation module and suggestion
module. The accumulation of data requires us to do the observation and analysis at the specific
amount of time. The observation continues at the specific point (based on dynamic distance)
before the merging point.
The first module is the simulation application development. This is the simulation part whereby
it is module that creates the simulation including the drawing and the simulation. By integrating
the result data and algorithm that we have explored, the accumulated data will be tested as a
set of data input during the application execution.
The second module represents the suggestion tool, as a result of data processing during
simulation application execution. Suggested tools comprises of several aspects of
recommended action that will be presented through output device, such as digital board at
highways. This tool will acknowledge the suggested speed for the drivers in order to enhance
the efficiency of travel time.
11. TASK DIVISION
•Coordinating the project flow
•Conduct a discussion regularly
MUHAMMAD AMMAR •Data accumulation of traffic observation
FIRDAUS •Implementation phase : main program
coding
•Analyzing the Kinematic Wave Theory
•Analyzing the Fluid Model Algorithm
•Data analysis
IBNU BUKHARIE •Implementation phase : data and algorithm
integration coding
•Quality assurance
Figure 4: Task Division
12
12. METHODOLOGY
Study the traffic flow at the merging point
Observation of traffic at a specific time to gather data
based on the parameter
Study the kinematic wave theory and fluid dynamic
algorithm
Develop a traffic simulation tool based on the data
collection by applying the enhanced algorithm
Conduct testing to prove the efficiency of the system
to reduce traffic congestion at merging point
Figure 5: Methodology of Traffic Simulation Project
Further studies in this topic have led us in defining the methodology or step taken in
implementing the project. The first stage of all is to study the traffic flow at the merging point.
This is to define the patent of traffic flow, as we are using the information in planning for the
next step.
Moving on, the next step is the observation of traffic at a specific time to accumulate data
based on the parameter. Obviously, the result of the observation is differing based on certain
condition, such as the observation during peak hours, raining and temporary obstacles
(accidents, lane closed, and road maintenance).
The third step in the development process is to study the kinematic wave theory and fluid
dynamic algorithm. This is the backbone of the simulation application because this algorithm
will determine the best recommendation to enhance the traffic flow.
The next step is to develop a traffic simulation tool based on the data collection by integrating
the enhanced algorithm into the application. This is the crucial step where the programming
part will be taken into action, in which all the related components will be bind into the codes.
This will result in producing the executable file as a standalone application.
13
The final step of this development is to conduct testing to prove the efficiency of the system to
reduce traffic congestion at merging point. The application that has been conducted will be
tested with the input data (based from data accumulation) to determine the dynamicity of the
application to suit with different parameter (condition).
13. CONCLUSION AND FUTURE WORK
This project proposes a suggestion tool in an efficient way to overcome traffic congestion,
especially at merging point. Merging point is where more than one lane with different paths is
merged into one path. There are many factors that causes in traffic congestion, depending on
the situation. By studying and applying the combination of fluid dynamic algorithm and
kinematic wave theory, it can improvise the traffic flow, thus the traffic congestion especially
during peak hours and festive seasons. We focus at the merging point because it is the main
causes of traffic congestion at Penang Bridge. This dynamic traffic simulation tool is defined
based on adjustable parameter values (distance, speed limit, vehicle type ratio), and these
parameter produces a different result, based on the current data input on the specific time. The
process of developing the system consists of the aspects of parameters, vehicles and road
design, algorithm integration and the verification of the dynamic simulation tool. The
simulation application gives a clear picture of the real world situation, based on the data
analysis and processing. Moreover, this application will run as a standalone application which
will produce an output of a suggestion speed when the data is entered dynamically. The future
work will cover the detail planning of the project including risks identification and mitigation
and methods of execution.
14
14. REFERENCES
[1] [Link]
[2] [Link]
[3] U.S federal Highway Administration, “Traffic Congestion and Reliability: Trends and
Advanced Strategies for Congestion Mitigation. “, Sept. 2005.
[4] Bretti, G. Natalini, R., and Picolli, B., “Numerical algorithms for simulations of a traffic model
on road networks”, J. Comput. Appl. Math., 2007,210,(1-2), pp. 71-77.
[5] Gabriella Betti, “A Fluid-dynamic Model for Traffic Flow on a Road Network”, Istituto per le
Applicazioni del Calcolo, M. Picone (IAC), March 2005.
[6] A. Carlos F. Daganzo, “A Simple Traffic Analysis Procedure”. Symposium on Networks and
Spatial, March 2001.
[7] M. J. Lighthill, G. B. Whitham: “On Kinematic Waves. I. Flood Movement in Long Rivers”,
Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol.
229, No. 1178, pp.281-316.
[8] Wen-Long Jin, “A Kinematic Wave Theory of Lane-changing Vehicular Traffic”, Institude id
transportation Studies, University of California, February 2008.
15