ADMM with SUSLM for Electric Vehicle Routing Problem with Simultaneous Pickup and Delivery and Time Windows

This paper studies the electric vehicle routing problem with simultaneous pickup and delivery and time window (EVRPTWSPD). In this paper, a novel alternating direction multiplier method with sequential updating scheme of lagrangian multiplier is proposed to optimize EVRPTWSPD with the goal of minimizing cost. This method first decomposes the problem into a series of augmented lagrangian submodels by using augmented lagrangian decomposition technology, then solves the submodel through labelling-setting algorithm. Finally this method iteratively updates the subproblems and lagrangian multiplier through sequential updating scheme of lagrangian multiplier. In experiment benchmark verification, the method proposed in this paper has excellent performance, and the method can get the tight lower bound in the process of solving.