Online joint optimization of pick order assignment and pick pod selection in robotic mobile fulfillment systems

Abstract

Robotic mobile fulfillment systems is a large dynamic system where information such as backlog, inventory, pod location, and robot state will change with time during its operation. In this paper, we model the dynamic system for the online joint optimization of pick order assignment and pick pod selection considering multiple items and then propose a new MIP model to solve the decision-making in the key process. We design a heuristic based on alternate decision and order-driven to solve large-scale problems. We extend an open-source simulation framework to evaluate the results of proposed model and algorithm through different instances in the dynamic system. Experimental results show that the performance of the new MIP model is at least 13% higher than that of the different settings, and the performance of the designed heuristic is at least 57% higher than that of the state-of-the-art sequential method in the literature.