THE VEHICLE ROUTING PROBLEM WITH TIME-VARYING TRAVEL TIMES AND A SOLUTION METHOD

A variant of the vehicle routing problem, which considers time-varying travel times between locations, has many real-life applications. It is shown that the problem cannot be solved to optimality even for very small instances. With respect to the additional computational efforts, the benefits gained in tenors of reduction in total routing cost through considering the time-varying property of the travel tunes is investigated. For this end, a genetic algorithm is proposed and experimental tests are conducted. It is found that although the time-varying property significantly increases the computational complexity, careful choice of constraints handling method and replacement strategy can improve the performance of the algorithm. It is also shown that the benefits gained in terms of objective value through considering the time-varying property of travel times is significant.