Research On Multi-objective Master Production Scheduling For Discrete Manufacturing Enterprises

Discrete manufacturing enterprises have the characteristics of complex product structure,flexible process route,limited production capacity,uneven balance of product requirements,and conflicting performance management target,and because of this,the plan,organization and coordination of the manufacturing process are often difficult.The master production scheduling plays an extremely important role in the production management of discrete manufacturing enterprises,which is the core hub of the enterprise production planning system,transforming the strategic plan defined in the production plan into tactical execution.The master production scheduling is called the tool of the senior manager to control the manufacturing resources,which is the main input source of the downstream material demand plan and the ability demand plan,and determines the production and supply of all product parts and materials of the enterprise.In the enterprise resource planning system,the master production scheduling coordinates the internal resources and market requirements of the company,which is the link between the strategic planning and the production scheduling execution,also acting as a connecting link between the preceding and the following and a transition between the macro and the micro.Because of the importance of the master production scheduling in the production planning system,the effectiveness and actuality of the master production scheduling must be guaranteed,otherwise the company may not respond to the needs of the customer or wasting resources.Based on the analysis of the present situation of the system of the theory,method and application of the production plan,the paper studies the model and method of the production plan management of the discrete manufacturing business owners and the research content of this paper mainly includes the following four aspects.(1)The connotation of the discrete manufacturing enterprise and its master production scheduling is introduced,and the main points of the input parameters,output parameters and target parameters of the master production scheduling are analyzed.Based on the problems of master production scheduling,material demand planning and capacity demand planning,a multi-objective integrated production plan model is proposed,and the conceptual model,process model and manufacturing list model of integrated production plan are constructed,and the logistics and process relationship of the integrated production plan,the operation logic,the manufacturing system,and the advantages of the integrated production plan are expounded.(2)In order to solve the problem of balanced use of production capacity in the production scheduling,this paper established single product,multi-stage and multi-objective master production scheduling model based on equilibrium production,considering not only net demand and production capacity constraints but also four performance management goals including the balanced production,on time delivery,inventory carrying and overtime production.Based on the mathematical characteristics of multi-objective nonlinear integer programming,auto-tuning strategy genetic algorithm(AT-GA)is designed for calculation.This algorithm designs an encoding method to encode the integer variable chromosomes to adapt the constraints of the model.The ideal point method is used to handle the four nonlinear goals of the model to obtain the adaptive value function.Using fuzzy logic control technology(FLC)from the change of the dynamic adjustment and the variation operator,the calculation ability of the genetic algorithm is increased by the global and the bureau's search ability of the equilibrium algorithm.Finally,two algorithm experiments are conducted to verify the ability of the multi-objective and the algorithm search ability,and sensitivity analysis of the model parameters is carried out.(3)In order to solve the problem of the lack of optimization mechanism of master production scheduling due to the limitation of resources and production capacity,this paper proposes a multi-product,multi-stage and multi-objective master production plan model under the dual constraint of resources and materials.The model aims at timely delivery,inventory reduction,overtime production reduction and safety inventory maintenance.According to the characteristics of the model,duality theory-based adaptive particle swarm optimization(DTA-PSO)was designed.The particle structure was defined by hierarchical integer coding,and the dual updating mechanism was used to replace the traditional updating mechanism.The concept of positive and negative elements is introduced,and each element in the particle is divided into positive and negative elements.Based on update probability,positive and negative elements are updated in pairs.This mechanism effectively guarantees the feasibility of updated particles and largely avoids invalid search,thus improving the search ability of the algorithm.The adaptive parameter mechanism is introduced,and the machine can adjust the algorithm parameters according to the number of iterations,and thus effectively balance the algorithm's global and internal search ability.Compared with lingo,the experiment found that when the problem complexity was too high,the lingo was unable to be effectively solved,while the DTA-PSO solution time would not be unstable as the scale of the problem expanded.Compared with the traditional particle swarm algorithm(PSO),it is better than the traditional PSO in the adaptive value index and the convergence algebra index,and verifies the advantages of the algorithm in solving nonlinear integer programming.Finally,sensitivity analysis of the model parameters is carried out.(4)In order to solve the problem of the master production scheduling in the demand uncertainty environment,this paper takes each product gross requirements as the uncertain variables,considering the uncertain characteristics of the subjective and the objective,using triangular fuzzy random Numbers to measure,at the same time,considering the inventory levels,product did not meet the requirements,the product is lower than the safety stock level and resource overload capacity of four goals,set up with a master production scheduling model of fuzzy random variables.Based on the characteristics of fuzzy random variable,multi-objective function and integer decision variable,the fuzzy random simulation-integrated adaptive particle swarm optimization with multiple structures(FRS-AMPSO)is designed by an integrated fuzzy stochastic simulation,which uses the binary encoding of the 0-1 variable to adapt to the discrete characteristics of the model decision variable.The fuzzy stochastic simulation technique is integrated into the process of particle performance evaluation,which effectively solves the problem of the uncertain variable in the model.With the multi-structure particle renewal mechanism,the opportunity of the sociology study of particles is extended,which improves the local search ability of the algorithm later.The adaptive variation operator and the inertial weight improvement scheme are introduced to improve the detection ability of the algorithm.Through the scenario simulation of the proposed model and algorithm for two different scale master production planning problems,the performance of FRS-AMPSO in dealing with the actual demand uncertainty multi-objective master production planning problem is verified respectively.Finally,sensitivity analysis of the model parameters is carried out.