Intelligent transportation management solution

Abstract

Intelligent Transportation Management Solution is a proof of concept project that aims to simulate an on-demand land transportation system whereby passengers are to be taken from one place to another. The solution is the result of research in the field of passenger transportation systems where unlike traditional bus transit systems, the system has no fixed routes and no fixed timetables. All user transportation requests are stored for the application to attempt to satisfy the requests by generating different routes and timetables depending on passenger demand. This thesis attempts to solve a variant of the Dial-a-Ride problem. The problem is a known NP-Hard problem whereby a number of user requests are inputted to an algorithm and the output is a set of routes for a number of vehicles. A user request consists of a user's pickup location (origin), his/her drop off location (destination) as well as the desired pickup time. This thesis explores the Vehicle Routing Problem and request clustering problems such that feasible solutions are obtained for each vehicle in the scenario. The system will abide by a number of constraints including the passengers' waiting time prior to pickup and the time duration a passenger may be on the vehicle. These constraints are all variable inputs to the system that may be varied. This allows a transportation company to use such a system to find the correct balance between a financially viable service and satisfying a certain degree of passenger comfort level. The output of the Intelligent Transportation Management Solution algorithm can then be fed into a simulator that graphically displays the routes, requests and vehicles. The project also sees the development of a mobile application capable of generating requests from passengers.