Production Planning And Scheduling Method Research For The Flexible Job-shop

Flexible job-shop have the characteristics of varieties,small batch production,complex production process.However,the existing MRP,ERP systems can not effectively solve the problem of production planning and scheduling in the environment.Therefore,the artificial planning and scheduling method is still used in the workshop of aerospace companies.But,the manual planning and scheduling method is difficult to fully consider so many constraints,this approach is hard to meet the needs of flexible production,and often led to the disjointed implementation owing to unreasonable production plan.Currently,the unreasonable production planning and scheduling become the key problem which restricts the workshop production efficiency to be further improved.Therefore,this paper investigates and analyzes the production process of the flexible job-shop,and researches around the problem model and the optimization method.The main research work of this paper is as follows:1)In order to solve the problem of modeling complex productionplanning problems,a modeling method for production planning problem of flexible manufacturing workshop is proposed,which is based on formal language of HTN.In this model,each variable in the model is mapped to a field in the database table.The model can be instantiated according to the material structure information,production process information,process step information and work equipment information.The initialed model can describe detailed production process characteristics.2)In order to solve the optimization problem of production task planning model,a production task planning method based on HTN and dynamic critical path method is proposed.This method can achieve the balance of production as the optimization objective.Based on the idea of continuously decomposing the complex problems and adjusting the non-critical path tasks and dynamically adjusting the critical path,the method realizes the production capacity balance.Finally,the production scheduling algorithm based on HTN-CPM is compared with the MRP task planning algorithm for an aerospace production workshop,and the results show that for the 4-month task planning problem,the proposed method results can reduce 1165 hours on average compared with MRP results,which proves that the proposed method can effectively optimize the production load.Compared with the MRP,the task scheduling algorithm has the advantage of balancing the production capacity indifferent time intervals.3)A single-objective and multi-resource constrained production scheduling model is established,which is a MINLP programming problem,based on the actual production of flexible job-shop.For the problem that the Boolean logic function in the MINLP scheduling problem makes the planning model discontinuous and non-differentiable,a Boolean continuous function based on Gaussian function is proposed.Based on the idea of limit approximation,Boolean logic function is replaced by the evolved Gaussian function,and then uses the penalty function method to transform the problem into a continuous smooth nonlinear programming(CSNLP)problem with simple boundary constraints.In order to solve the optimization problem of CSNLP,a DE-LBFGSB hybrid global optimization algorithm is proposed.In the case analysis,the DE-LBFGSB hybrid algorithm is combined with the GA algorithm,PSO algorithm,DE algorithm and their improved algorithms.The results show that the proposed algorithm can reach the late convergence state within 400 generations,and the solution quality is superior to the PSO algorithm,DE algorithm,JDE algorithm and MDE_PBX in the 4000 generation of optimization results,proved the proposed hybrid algorithm has a significant advantage in convergence speed,and it can converge quickly to the satisfactory solution.Moreover,the proposed algorithm has a very good performance in globaloptimization.