Research On The Robust Scheduling Of Multi-Variety And Multi-Process And Multi-Unit Manufacturing System Based On Improved Cellular Automata

As the global markets compete fiercely, manufacturing enterprise need fast response to the market, the production process of product have to be adjusted continually, according to the customers’requirement, to personalize the customization, and therefore the manufacturing mode of multi-variety and multi-process and multi-unit become mainstream mode. This mode of production system is a completely discrete and complicated system with many varieties, small batch and unit layout. Along with the influence of uncertainties, robust scheduling is considered as important research direction. However the existed solutions can’t get a satisfactory answer. Robust scheduling problem in multi-variety and multi-process and multi-unit manufacturing system is a typical NP-hard problem. As modeling tools which can solve complicated problem, Cellular Automata and Genetic Algorithm both have their own advantages for such discrete systems. So it is to some extent valuable and feasible to model, simulate, analyze and optimize the robust scheduling by combining the two methods.According to the characteristics of Cellular Automata and Genetic Algorithm, a way of combing the two methods is put forward to apply in the robust scheduling problem in multi-variety and multi-process and multi-unit manufacturing system. The major contents are as follow:1. Study on the original characteristic of the robust scheduling problem in multi-variety and multi-process and multi-unit manufacturing system, summarize its features and abstract the shop scheduling module.2. A primary Cellular Automata structure is set up to describe the robust scheduling problem in multi-variety and multi-process and multi-unit manufacturing system. Considered the structure features of Cellular Automata, the robust scheduling system is described as a two-dimension network space, designed the station as cell, the work piece as moving particle, the takt time as initial condition and boundary condition, the scheduling rules as local self-evolution rules, to carry out the mesh mode of Cellular Automata.3. Used Genetic Algorithm to optimize local self-evolution rules of Cellular Automata, set up the robust scheduling of multi-variety and multi-process and multi-unit manufacturing system based on improved Cellular Automata, so as to acquire the optimal scheduling solution. 4. Based on Microsoft Visual C++and Oracle database mixing programming way to realize the simulation system of robust scheduling of multi-variety and multi-process and multi-unit manufacturing system. An example will verify the validity and feasibility of the scheduling algorithm.5. Empirical research of PTCN production workshop scheduling, simulated the production scheduling, obtained the optimal scheduling solution of the workstation group. Contrasted practical scheme and optimal scheme, optimal scheme reached the better result, proving the feasibility and effectiveness of the model and method.6. In the future, the consideration of more kinds of uncertainties and re-scheduling scheme could be further.