Research On The Uncertain Integrated Process Planning And Scheduling Problem Based On Interval Theory

Integration of process planning and scheduling (IPPS) can effectively optimize themanufacturing result. Process planning defines how the product will be processed consideringits features and technical factors. Scheduling is one manufacturing function that attempts toassign manufacturing resources to the operations according to the process plans. And somerelevant criteria, such as makespan, due dates are satisfied after scheduling.Conventionally, process planning and scheduling are two isolated manufacturingfunctions and this may lead to ineffective resource utilization and inflexible scheduling. Someresearchers concluded the benefits of IPPS which include reduction of human intervention,resource confliction and increase of machine utilization. Since it is an NP-hard problem andhas complex constraints, it is difficult to search for the optimal solution in the huge solutionspace. IPPS is a hotly researched topic to provide a blueprint of efficient manufacturingprocess.But IPPS in fuzzy environment is rarely researched because of its complexity. Thisresearch target on the uncertain IPPS problem, take external uncertainty into consideration.Those uncertain events include uncertain transportation time, setup time, tool loading andunloading time, machine breakdown and so on. The uncertainty increases the complexity offuzzy IPPS, so there is few works on this.Traditionally, uncertain processing time is modeled as triangular fuzzy number ortrapezoidal fuzzy number. Interval number is the simplest form of representing uncertainty inthe decision matrix. A minimum amount of information about the values of attributes isrequired by interval number. So in this study, interval number is used as representation offuzzy processing time. Based on the interval theory, a new probability and preference-ratiobased interval ranking method is proposed. The new ranking method has the simplecomputational process and can precisely rank interval numbers. The mathematical model ofuncertain IPPS based on interval theory is proposed.Based on the features of uncertain IPPS, Genetic Algorithm (GA) is used to optimizeIPPS with interval processing time. In order to effectively solve this problem, the integratedcoding method is adopted in process planning system. The operation-based coding method isused in scheduling part. Efficient genetic operators including crossover and mutation aredesigned. Experiment results show the effectiveness of GA.In order to obtain better results in large scale problem, one hybridized algorithm, Particle swarm optimization (PSO) with genetic operators is used. To improve the search capability ofthe hybridized algorithm the special operators are used. PSO is redefined in the hybridizedalgorithm. The learning strategy of particles is kept and the update of generations obtaingenetic operators. Some strategies are designed to prevent the particles trapping into localminimum. The results generated by the proposed algorithm illustrate the effectiveness.