Hyper-Heuristic Coevolution of Machine Assignment and Job Sequencing Rules for Multi-Objective Dynamic Flexible Job Shop Scheduling

Abstract

Nowadays, real-time scheduling is one of the key issues in cyber-physical system. In real production, dispatching rules are frequently used to react to disruptions. However, the man-made rules have strong problem relevance, and the quality of results depends on the problem itself. The motivation of this paper is to generate effective scheduling policies (SPs) through off-line learning and to implement the evolved SPs online for fast application. Thus, the dynamic scheduling effectiveness can be achieved, and it will save the cost of expertise and facilitate large-scale applications. Three types of hyper-heuristic methods were proposed in this paper for coevolution of the machine assignment rules and job sequencing rules to solve the multi-objective dynamic flexible job shop scheduling problem, including the multi-objective cooperative coevolution genetic programming with two sub-populations, the multi-objective genetic programming with two sub-trees, and the multi-objective genetic expression programming with two chromosomes. Both the training and testing results demonstrate that the CCGP-NSGAII method is more competitive than other evolutionary approaches. To investigate the generalization performance of the evolved SPs, the nondominated SPs were applied to both the training and testing scenarios to compare with the 320 types of man-made SPs. The results reveal that the evolved SPs can discover more useful heuristics and behave more competitive than the man-made SPs in more complex scheduling scenarios. It also demonstrates that the evolved SPs have a strong generalization performance to be reused in new unobserved scheduling scenarios. INDEX TERMS Scheduling, flexible job shop, hyper-heuristic, multi-objective, genetic programming. NOMENCLATURE MO-DFJSP multi-objective dynamic flexible job shop scheduling problem MAR machine assignment rule JSR job sequencing rule SP scheduling policy GEP genetic expression programming CCGP cooperative coevolution genetic programming with two sub-populations TTGP genetic programming with single population that a GP individual contains two sub-trees NSGAII nondominated sorting genetic algorithm II SPEA2 strength Pareto evolutionary algorithm 2

Dynamic Scheduling Framework of the Flexible Mixed-Model Assembly Line Based on the Internet of Manufacturing Things

“…The dynamic production scheduling (DPS) [12]- [20] means that the planned scheduling solution can be adjusted in real time to minimize the impact on the performance of production system when disruption events occur randomly. Furthermore, the adjusted scheduling solutions should be consistent, efficient and robust to complete production tasks against disruption events.…”

“…This paper here proposes a weighted sum of properties (WSP) dispatching rule, as shown in Equation (17). On the basis of hyper-heuristic concept [15], [16], [20], the WSP dispatching rule needs the IoMT-based FMMAL actively perceive the production conditions and dynamically search the priority weights to generate the priorities according to the optimization objectives. Three considered process properties of products in MC are the assembly time, due time and assembly power, which correspond to three optimization objectives of minimizing the C max , TET and TEC.…”

“…The Chromosome 1 is used to search for the planned product sequence. It adopts the sequential encoding and the genes represent the planned product codes in the source, as shown in Equation (20).…”

Dynamic Scheduling Framework of the Flexible Mixed-Model Assembly Line Based on the Internet of Manufacturing Things

“…Zhang and Wang [19] employed the constraint programming and mixed-integer linear programming to formulate the flexible assembly job-shop scheduling problem in a dynamic manufacturing environment, and developed several dispatching rules with machine feedback mechanism. Zhou et al [20] proposed three types of hyper-heuristic methods for coevolution of the machine assignment rules and job sequencing rules to solve the multiobjective dynamic flexible job shop scheduling problem, and found that the heuristics discovered by the evolved scheduling policies are more useful and competitive. The above studies are very inspiring, but they fail to design the hyper-heuristics on the basis of product properties, including the assembly time, due time and assembly power.…”

A Hybrid Machine Learning and Population Knowledge Mining Method to Minimize Makespan and Total Tardiness of Multi-Variety Products

“…Ahmadi et al [16] used NSGA-II and non-dominated ranking genetic algorithm (NRGA) to optimize the makespan and stability under the disturbance of machine breakdown. Zhou et al [17] presented a cooperative coevolution genetic programming with two sub-populations to generate scheduling policies. Zhang et al [18] utilized a genetic algorithm combined with enhanced local search to solve the energy-efficiency job shop scheduling problems.…”