Abstract: The rapid growth of logistics distribution demand makes an increasing wide application of pickup and delivery. A pickup and delivery problem with time windows and weight-related cost is studied, considering that the cargo weight has an effect on the transportation cost in the distribution process. A bi-objective mixed integer programming model of the problem is formulated to minimize the number of vehicles and the total transportation cost, in which the transportation cost is a function of vehicles’ load capacities and travel distances. A two-stage adaptive large neighborhood search algorithm is designed to solve the model, where the performance is improved by designing multiple efficient destroy and repair operators and introducing a simulated annealing strategy to avoid local optimal. Instances with different scales and characteristics are tested, showing that the proposed two-stage adaptive large neighborhood search algorithm can solve small-scale, middle-scale and large-scale instances effectively. Also, the impact of cargo weight and different transport coefficients on transportation cost is analyzed, providing reference for logistics enterprises to optimize the route of pickup and delivery vehicles.

Key words: pickup and delivery problem, weight-related cost, time window, adaptive large neighborhood search algorithm