Job Shop Scheduling with AGV Charging Considering Travel Time and Processing Time

Abstract

Job Shop Scheduling Problem with Automated Guided Vehicles (JSPAGV) better aligns with the real-world workshop scenarios and has become a research hotspot. However, optimizing JSP-AGV with AGV charging remains a significant challenge. The comprehensive JSP-AGV model incorporated AGV charging is established, where charging is mandatory and interrupts the transportation task. Then, an improved genetic algorithm with a hybrid initialization strategy and problemspecific crossover and mutation operators is devised to minimize the Exit Time (ET). Extensive simulations are conducted to evaluate the model and algorithm. Furthermore, Design of Experiments (DoE) is employed to quantitatively analyze the impact of three critical system parameters, such as AGV battery capacity, AGV quantity, and travel time-to-processing time ratio. The analysis reveals that the travel time to processing time ratio determines the scheduling bottlenecks in JSP-AGV. Orthogonal experimental results indicate that for time-related metrics, including ET, machines waiting time for jobs, jobs waiting time for machines, AGVs waiting time for jobs, jobs waiting time for AGVs, the influencing factors are ranked in descending order of importance as follows: travel time to processing time ratio, AGV quantity and AGV battery capacity. In contrast, for AGV charging frequency metric, the order of the influence is AGV battery capacity, travel time to processing time ratio and AGV quantity.OPEN ACCESS Received: 13/11/2025 Accepted: 04/02/2026 Published: 29/05/2026

Document