Traffic-flow model

No Thumbnail Available
Herbert, Nicholas
Issue Date
BSc in Computer Science
Dublin Business School
Items in Esource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
This thesis is concerned with the implementation of a traffic model and effecting changes in the model in order to gain insight into traffic flow and traffic control in a transportation system or network. Various performance measures produced from the traffic model can be analysed for this purpose, these include the averages and standard deviations of system variables such as the waiting time for cars, queue lengths at traffic lights, total time vehicles spend in the system and the total queue length. State of- the-art ideas and concepts in the field of transportation theory and transportation science are examined in this document. The main functions of traffic-flow models are deemed to be: 1) The evaluation of innovative transportation management policies, designs and concepts. 2) Cutting time and cost for evaluation and design of transportation concepts. 3) Reducing the risk and disruption to traffic-flow caused by experimentation in the field. 4) Providing better evaluation of design and operational improvements. 5) Guaging the operations of transportation facilities and transportation systems. The actual problem of traffic-control is considered objectively in this thesis and justification is given for how the research undertaken in this thesis reinforces ideas already put in place by others, these include the fact that traffic-models are useful for allowing "what-if' scenarios and avoid the time, disruption and cost of field experimentations, also the fact that a simulation can be repeated as many times as a person wishes with different parameters to get the true essence of the physical system being simulated. Reasons are given as to why this problem merits research in the fact that it has such an affect on a great diversity of people and quality of their lives today. Conclusions and inferences drawn from the research undertaken into the problem of traffic control in a network is that the traffic-model implemented accurately simulates traffic-flow in a road-network and allows for the optimization of different transportation policies including the setting of traffic-lights at junctions by examining statistics such as the average waiting time of cars in queues at traffic-lights. The exponential distribution seemed to be the most suitable distribution to simulate the arrival times of cars into the road-network, also that the problem of traffic-control is a worthwhile problem in solving in that it affects so many people.