An Optimal Bus Scheduling Model Based on Mixed-Integer Linear Programming

The bus company mainly relies on the experienced scheduling personnel to prepare the time plate of the bus line manually in advance, and then conFigure the vehicles. However, this kind of scheduling method often cause a waste of bus seat resources only by experience. In this paper, we propose a vehicles scheduling model called MILPS (Mixed-Integer Linear Programming Model of Scheduling) to minimise the gap between the demand and supply of bus seats. In the first phase, we first analyze the historical trajectory data of taxis in Xiamen and select the two most popular stations as the starting and ending points of bus routes, then generate the service area between them by rectangle. In the second phase, we derive several criteria to build bus scheduling model based on Mixed-Integer Linear Programming. Finally, we use CPLEX optimizer to solve this MILP problem. Experimental results show that the proposed scheme can effectively reduce the waste of bus seat resources (>76.8% reduction) and significantly improve the delivery ratio (>71.5%).