Study On The Method Of Disruption Management Scheduling For Hybrid No-wait Flow Shop

In recent years, with the development of information science technology and the transformation of economic development mode, customer requirements become more and more diversified and personalized, which leads to more fierce market competition and more complex manufacturing process. Both of the changes in external and internal environment set forth stricter requirements for production scheduling of manufacturing enterprises. Manufacturing enterprises facing the issue should strengthen scheduling schemes’quick responsiveness to uncertain events as far as possible in production scheduling, especially those with complex production process. Meanwhile, they should shorten the manufacturing cycle of products, improve the utilization of equipment, reduce the material and energy consumption in production process, ensure the punctual production, and improve customer satisfaction and so on.Hybrid no-wait flow shop is very common in steel, petrochemical, pharmaceutical and other process industries. To solve the initial optimal scheduling infeasible problem caused by various kinds of uncertain events in hybrid no-wait flow shop, disruption management scheduling methods are studied in this paper. Firstly, aiming at the situation with concurrent disruptions, a disruption management model with two objectives is put forward:the original one (min. total weighted completed time) and the disruption fixing one (min. total weighted delayed time). Based on a multi-objective random weighted policy supposed by dynamic changing search directions, a hybrid algorithm combined improved local search mechanism with Particle Swarm Optimization is proposed. Secondly, another disruption management model, which considering a JIT delivery satisfaction based behavioral and a machine allocation deviation reflected consistency of the schedules as two disruption fixing objectives, and still take min. total weighted completed time as the original objective, is also put forward. And to this disruption management model, a PVPS algorithm combined global rapid optimum of PSO and local search superiority of VNS is proposed. Finally, numerical experiments are designed separately to verify the algorithms. Simulation data show that, not matter the completely objective disruption management scheduling method, but also the disruption management scheduling method considering behavioral factors are all effective. Hybrid no-wait flow shop is a relatively complex production process. The disruption management scheduling methods proposed in this paper could provide guidance to iron and steel, textile and other continuous type of complex production process for their practical scheduling, and also provide reference for research of production scheduling problem in operational management.