Model Design of Job Shop Scheduling Based on Multi-agent System

Job-Shop Scheduling is the core contents of the CIMS research area and production logistics in the discrete manufacturing industry. To solve Job-Shop Scheduling problems, a scheduling model based on MAS (Multi-Agent-System) is presented with multiple applications of genetic algorithm (GA) and contract-net mechanism. In the model, the entities in the job-shop are abstracted as Task Agent, Resource Agent, Control Agent and Cooperation agent. The scheduling model is able to combine the flexibility of MAS and optimization through algorithms mentioned above. This model improves the literature proposed genetic algorithm, solves the dynamic job-shop scheduling problem by multi-agent consultative mechanism. According to compute the stimulating degree and equipment threshold, the new mission can be directly access to the appropriate sequence. At the same time, the use of genetic algorithm processing of the final route selection makes a reasonable allocation of resources. At the same time, the scheduling optimization objectives and the factors disturbing job-shop scheduling are analyzed in the paper.