A Hybrid MIP/Heuristic Model for Experience Based Driver Assignment

Abstract

In this paper, we describe an interesting driver assignment problem that is computationally intensive to solve due to its combinatorial nature. A hybrid approach involving mixed integer programming (MIP) and a heuristic is used to give good solutions to the problem within reasonable computation time. This approach attempts to utilize the strengths of MIP to search for an optimal solution, while letting the heuristic component address the complexity involved in the driver assignment problem so as to improve the time required to obtain a solution. Computational results are used to illustrate the performance of the approach