Scheduling the Just-in-Time Delivery of Parts for Mixed-Model Assembly Lines Considering the Electrical Energy Consumption of an Automated Guided Vehicle Trolley

Abstract

This paper studies the scheduling of just-in-time part delivery for a mixed-model assembly line with material supermarkets under an unfixed path. According to the problem description and related assumptions, a mixed-integer programming model with the goal of minimizing the line-side inventory and delivery energy consumption of automated guided vehicles (AGVs) was constructed. Based on the material demand calculated according to the assembly plan, the model decides the parts to be delivered, the schedule of each delivery round, and the route of the AGV trolley. For small-scale problems, the epsilon constraint method is used to solve the Pareto frontier accurately. For large-scale problems, an improved version of the NSGA-II (Non-dominated Sorting Genetic Algorithm II) is designed to solve the problem more efficiently. Numerical analyses of different scales demonstrated the efficiency of the proposed algorithm. Furthermore, a sensitivity analysis showed that using a smaller AGV capacity with more frequent delivery can achieve a lower inventory level but higher energy consumption. Therefore, enterprises need to choose reasonable delivery schemes according to the assembly line layout and station occupation to save assembly space and achieve the target of carbon emissions reduction.