|
||
| Home | Islamic Microfinance | Vedanta | Islam | ITMS | Organisational dynamics | Islamic Finance | PLADS | Indus Writing | My Resume | The Todas | ||
|
||
|
Dishaa |
||
|
||
|
ITMS |
||
|
||
|
INTELLIGENT TRAFFIC MANAGEMENT SYSTEM An inter-disciplinary paper on Decision
Science and Information Systems |
||
|
||
|
ABSTRACT Traffic congestion has been increasing world-wide as a result of increased motorization,
urbanization and population growth. Congestion reduces utilization of the transportation infrastructure and increases travel
time, air pollution and fuel consumption. With a limited road infrastructure that strives to keep up with the booming increase
in the number of commuters, the only way out of this disarray for developing countries like Keywords: traffic control, fuzzy logic, decision support INTRODUCTION: DECISION SUPPORT SYSTEMS are interactive computer-based systems which help decision makers utilize
data and models to solve unstructured problems. They couple the intellectual resources of the individual with the capabilities
of the computer to improve the quality of decisions. The benefit usually associated with a Decision Support System is
that it leads to a higher decision quality in terms of cost reduction, increased productivity and timesaving. The proposed system is one for the traffic commuter, the DSS is made for him to understand the scenario
based on the data and its analysis. Due to increasing complexity in the traffic scenario, trial and error methods in decision making may
no longer be sufficient. Hence the proposed system. The decision making process starts with the intelligence phase, where reality is examined and the problem
is identified and defined. Data collection is the major challenge in this phase. In the design phase, a model that represents
the system is constructed. This is done by identifying key variables and defining the relationships between them. The choice
phase includes a proposed solution to the model. Once the proposed solution seems to be reasonable, it is ready for the implementation
phase. Now, we examine the various phases with respect to our Traffic Management System. INTELLIGENCE PHASE : ●
Finding the problem ●
Problem classification ●
Problem Decomposition Problems arise of a dissatisfaction with the way things are going. The existence of a problem can be
established by monitoring and analysing the system. The current traffic scenario is one plagued by problems. Recurring time
delays due to traffic congestion and irregular flow through traffic bottlenecks cause much distress to the commuters. The classification of the problem is done according to the degree of structuredness evident in the
problem. Programmed problems are those which are well-structured and are repetitive. Standard mathematical models have been
shown to solve these problems. Nonprogrammed problems are novel and recurrent and poorly structured. Semistructured problems
fall between the two extremes. the Traffic scenario presents a problem that has predictable problems but involve a high degree
of uncertainty due to the lack of structure. We therefore classify them as nonprogrammed problems. Problem decomposition is the division of a very complex problem into subproblems. Solving
the subproblems leads to a solution to the larger problem. The problem at hand is decomposed into a number of subproblems,
each subproblem representing a small geographical area, with its boundaries clearly defined. The various areas are connected
through nodes, represented by intersections or traffic signals. The actual decomposition may be carried out in any way suitable
to the road network under study.
|
||
|
||
|
|
||
|
||
|
||
|
||
|
DESIGN PHASE: ●
Understanding the problem ●
Developing possible course of action ●
Modeling The problem as we formulate it is one of improper
decision making on the part of the commuter. In a congestion scenario, the commuter should avoid the 'area' that is infected
and chose alternate routes. Here the 'closest distance' algorithm should be avoided as the area which lies on the closest
distance is infected with congestion. In such a case, persistence on the closest route is meaningless and might lead to complications
in the scenario. The solution is to chose alternate routes, and the challenge in the design phase is to make that information
available to the commuter. Sensing Techniques are reviewed for actual infection
diagnosis and Communication techniques are reviewed to understand ways of transmitting the data. Sensing Techniques New sensing technologies have greatly enhanced what
can be done using Intelligent Transportation Systems. Sensing systems can be either infrastructure systems or vehicle based
systems. Infrastructure sensors tend to be devices that are installed in the road or around the road, usually as part of road
construction/maintenance, while vehicle sensors are those sensors that are found installed in a vehicle. Inductive loops can
be placed in a roadbed to detect vehicles as they pass over the loop by measuring the vehicle's magnetic field. The simplest
detectors simply count the number of vehicles during a unit of that pass over the loop, while more sophisticated sensors estimate
the speed, length and weight of vehicles and the distance between them. Loops can be placed in a single lane or across multiple
lanes, and they work with very slow or stopped vehicles as well as vehicles moving at high-speed. Traffic flow measurement using video cameras is another form of vehicle detection. Since video detection systems do not involve
installing any components directly into the road surface or roadbed, this type of system is known as a "non-intrusive" method
of traffic detection. Video from black-and-white or color cameras is fed into processors that analyze the changing characteristics of the video image as vehicles pass. The
typical output from the Image Processing System is lane-by-lane vehicle speeds, counts
and lane occupancy readings. Vehicular Communication Communication technologies are getting faster by the day. The Transport System can benefit greatly from the application
of information and communications technologies. Intelligent Transportation Systems vary in technologies applied, from basic
management systems such as car navigation, traffic light control systems, container management systems, variable message signs
or speed cameras to monitoring applications such as security CCTV systems. A simple Graphical User Interface can be an effective Vehicular Communication System tool. The GUI
should be intuitive and easy-to-use such that the driver's attention should not be disrupted when using the system. A reasonable
range of communication is equally important. The system must be able to operate reliably within a reasonable range that motorists
are expected to be in for a sustained period of time. The fast emerging Bluetooth technology can serve as
an adequate Short Range Communication System. Longer range communications have been proposed using infrastructure networks such as Global System for Mobile Communications.
Long range communications using these methods is well established, but these methods require an extensive infrastructure
beyond what is installed in a vehicle. Fuzzy Control The employment of Fuzzy Control is recommended for very complex processes, when there is no simple
mathematical model. It is obvious that traffic flow does not follow a uniform set of actions that can be precisely defined
by a mathematical model. The complexity of relationships in a traffic flow analysis system is not one that can be represented
analogically. The fuzzy model consists of 4 components. The Linguistic database contains the knowledge in the form
of 'if-then' rules which is stored in a fuzzy rule base. The rules are given precise mathematical meaning through user supplied
definitions of the linguistic variables. The Fuzzifier maps the crisp values into suitable fuzzy sets. The fuzzy inference engine interprets the fuzzy sets provided by the fuzzifier. It uses the fuzzy rule
base knowledge to produce some fuzzy sets in the output. The Defuzzifier converts the output fuzzy set to a standard signal. Based on theoretical analysis of traffic control, generalized fuzzy rules using linguistic variables
and validation of fuzzy control principles are carried out. The input parameters we are looking at are i.
the number of vehicles entering the street
under study (street A) ii. the number of vehicles leaving street A iii. Average time for a vehicle to cross the length of street A iv. neighbourhood variable (traffic situations in preceeding areas) The fuzzy model equation of the isolated street (street A) can be derived based on these parameters.
The basic fuction that will be constructed will apply to a simple one-way scenario. Later two way models and nodal connection
equations can be designed. Ideally a model should generate an optimal order level. Most Optimization techniques work only for
a structured problem. Our approach to the problem at hand is to break the problem into sub-problems that are structured enough
to apply optimization. By considering Street A as an isolated problem, we aim to solve micro-problems and then work towards
a unified solution. CHOICE PHASE: The fuzzy model described above will have input parameters of traffic variables and time variables.
At the basic level, this input can be fed into the fuzzy model and an output can be obtained. The Output is one value, which
has been defuzzified. The current model proposes a 4mode output signal. The output from the fuzzy model will cause one of
the four output modes to activate and that is the communication to the user. The 4 modes can be classified based on the traffic
scenario in the relevant area. For instance, if a one way one vehicle average trip time on Street A is 5 minutes, Signal 1 would be
activated if the inputs suggest that a vehicle is now taking 1-3 minutes to cross Street A. Signal 2 would be activated if
inputs suggest the travel time as 4-6 minutes. Similarly, Signal 3 would activate for 7-9 minutes and Signal 4 for a travel
time in excess of 10 minutes. Is the choice making now easier? Instead of relying on mere hearsay and intuition, the commuters have
accurate information on traffic scenarios across the areas they intend to travel. The proposed signals can be installed in
individual vehicles or after a more detailed analysis of the System, can be built into the traffic signal infrastructure,
where color codes could give you the status of the road ahead. In the basic model, the actual choice is to be made by the commuter. He can base his decision on the
basis on the data made available to him. He may choose to travel on a busy street if his successive travel areas are relatively
free. The basic model may also be enhanced using techniques described in the 'Improvements' section. IMPLEMENTATION PHASE: ● Feasibility ● Improvements ● Future Services Feasibility Analysis is studied for various aspects of the proposed system. Recent advances in vehicle electronics have lead to
a move toward fewer more capable computer processors on a vehicle. A typical vehicle in the early 2000s would have between
20 and 100 individual networked Programmable logic controller modules with non-real-time operating systems. The current trend
is toward fewer more costly microprocessor modules with hardware memory management and Real-Time Operating Systems. The new
embedded system platforms allow for more sophisticated software applications to be implemented thus enabling the use of Intelligent
Traffic Management Systems. CASE STUDIES : A study of such endeavours in the recent past shows methods that were employed. It also serves to show that information
systems have proved useful in real time situations. I.
TCS developed a rail management system that
was responsible for automatic tracking and routing of trains. The train related data and parameters were stored and retrieval
made easy. A simple GUI was provided for easy access to the data. II.
The City of Vancouver, Washington implemented
an adaptive control system for traffic signal operations at 12 intersections along Mill Plain Blvd. Performance measurement
of this system was the main objective of this research. Link, intersection, and travel-time data were compiled and statistically
analyzed. Data observed from travel-time runs and data collected from system detectors were used to compare performance of
the system. This research showed that adaptive traffic signal control generally has a positive impact on the system. Based
on the operational studies, average speed improved up to 25%, the travel time decreased up to 20%. III. NextBus is a vehicle tracking system for public transportation
vehicles, especially buses and trams/light rail operations. Each vehicle is outfitted with a GPS receiver, which transmits
data about bus or tram speed and location to produce projected arrival/departure times. These times are broadcasted to electronic
signs at bus stops and tram stops and to the Internet including to wireless devices such as cell phones and PDAs. NextBus
also provides management tools so that transit agencies can better manage their systems. NextBus claims this system increases
user satisfaction by increasing the amount of information regarding delays and other disruptances that reaches the average
commuter. This information can be distributed by the Internet and also by signs at bus and light rail stops. The system is
used on the San
Francisco Municipal Railway for 3 bus
lines and all of the light rail services. FUTURE SERVICES: The proposed ITMS can play a very important role in the years to come. If implemented,
it will benefit greatly from regular analysis and upgradation of the system. Even the very basic techniques described here
can be coupled with the right applications to provide a host of services that are vital for the transport system. The archived data mart can be used to analyse traffic
information for the department, which would then place in a better position to take steps to tackle the problems they face.
The data warehouse can be a tool to study trends that enable traffic prediction with accuracy. Intelligent traffic signs can
help reduce waiting time at intersections. En-route transit information
and interactive
traveller information will make the information system a traffic infoline. Traffic network monitoring ensures smooth flow of
traffic. Transit vehicle tracking with automatic number plate recognition will enhance security features of the proposed model. Environmental concerns can be addressed by checking pollution levels by emissions recording
and analysis. Variable speed limits can be defined in accordance with the area and enforcement of traffic rules can be made
easier. An 'area'-wise parking management system would facilitate the very complex
parking mechanism. The presence of a stand-by Road-Maintenance fleet can be utilised fully with the help of the ITMS. Pattern recognition can be used to predict accidents
and roadway warning systems with intersection collision warnings can be installed.
In addition, Emergency response management can benefit greatly with emergency vehicle routing in place. Disaster command and
control will be less chaotic if we employ the services of the ITMS. IMPROVEMENTS: 1.
Expert System is a system that can replicate the expertise
of a traffic expert which is stored as the knowledge base. The end users can call on the computer for specific advice as needed.
The Expert System can make inferences and arrive at specific conclusions. 2.
Neural computing may be used to find patterns in the
traffic congestion cases and arrive at conclusions of what causes the congestions. The implementation of the DSS
is the duty of the traffic control department. In addition to saving valuable time for the commuters, it also aids them in
their Data collection, and they can analyse it for their own ends. The proposed DSS will enhance their ability to enforce
security, track traffic movements that enable planning for the future in road transport and a host of other possible functions
that have been described. The costs involved in setting up the proposed system are easily justified by the magnitude of the
problem and the fact that there is little else we can do to solve the problem in an organised way. CONCLUSION: The proposed Intelligent Traffic
management system has been theoritically shown to be efficient in dealing with the current problems and feasible to implement.
Attempts at building real-time models and their testing should now be carried out and the Traffic Control Department should
consider using the suggested Information Services tools to enhance their own performance and to help them serve the people
in a more effective manner. REFERENCES: 1. 1.
Decision Support Systems and Intelligent Systems
Efraim Turban and Jay E. Aronson 2. 2.
http://www.trb.org/news/blurb_detail.asp?id=6906 3. 3.
http://lib.tkk.fi/Diss/2002/isbn9512257017/ 4. 4.
http://pubsindex.trb.org/document/view/default.asp?lbid=776184 5. 5.
http://www.cse.iitd.ernet.in/~rahul/thesis.html 6. 6.
www.tcs.com/0_case_studies/service_practices/atc_adv_computer.htm 7. 7.
http://en.wikipedia.org/wiki/NextBus 8.
|
||