Allocation of buses to networks with overlapping routes: a case study of city of Dar es Salaam.

This project addresses the optimization model for bus scheduling problem. In general, the problem of allocation can be defined in the following terms: Given the origin destination transit matrix for the bus trips for the design period and the underlying bus network characterised by the overlapping routes, how optimally to allocate the buses among these routes? The problem formulation is developed which recognizes passenger route choice behaviour, and seeks to minimize a func¬tion of passenger waiting time and crowding level of the passengers in the bus subject to constraints on the available number of buses and enough capacity to be provided on each route so as to carry all passengers who would select it. The bus scheduling problem is solved in two stages: In the first stage, the frequencies required to handle the maximum loaded link on each route are identified. In the second stage, the surplus allocation is illustrated for the simple objective of minimizing the maximum crowding level on any route. Tabu search algorithm has been used to solve the problem of scheduling of surplus fleet. The model has been applied to Dar es salaam city network and found that about 93.6% of the average waiting time obtained by SUMATRA have been minimized. This directly minimized the overcrowding of passengers at the bus stop, as a result the maximum crowding level of the passengers in the buses have been minimized.