Buffer Allocation Optimization For Multistage Flow Shop Based On Queuing Network Model

In modern industry make-to-order production enterprise plays a very important role. As the order production, the MTO enterprises in the production process is full of random factors, and thus cause many problems such as the production is not stable, delivery date rampant tardiness. However, the project delivery as one of the MTO enterprise core competitiveness and an effective method which based on Workload control (WLC) delivery forecast is used to manage it at present. But, in order to accurately predict delivery, you must first be able to accurately define the workload norms, while one of the most commonly used and most critical is the workshop buffer allocation. Therefore, An effective research of the Buffer Allocation Problem (BAP) and the optimization method is urgent and meaningful.Consider these random factors during the production process of the MTO, this paper combine with the theory of Workload control and queuing network, and have research on the BAP and optimization of the multi-stage flow shop. The concrete content as follows:First, this paper describes the methods and limitations of the existing workload norms setting, and through the analysis of the characteristics of a single type workpiece flow shop, in order to determine the relationship between the load delimitation and cache settings. Then analyzed and select the objective function of the BAP, and proposed two solutions for the difficulties.Secondly, it detailed analysis of the various states of the queuing network node, and lists the corresponding continuous-time Markov chain transfer rate balance equation. At the same time, it analyzed the steady-state method for solving the CTMC state transition rate balance equation and derived the solution of the equation from the corresponding model. Afterward, developed a program in the MATLAB environment to solve the equation of state, and a numerical example is compared in order to prove its effectiveness.Then, in order to extend the solution of BAP to large, complex manufacturing systems, this paper used the simulation software of Tecnomatix Plant Simulation8.2to construct a simulation model which contained a flow shop with eight workstations, and designed a heuristic algorithm for optimizing the buffer allocation of the model. Comparing the calculation results in the literature by the simulation results and proved that the use of the simulation method is an effective means to solving the Buffer Allocation Problem.Finally, by numerical calculation and simulation statistical methods this paper design a series of examples of the objective function for buffers allocating, then detailed analysis and authenticated the system performance.In this paper, using the queuing network and workload control theory, built a flow shop buffer allocation model and optimized the model in a random environment, and presented the corresponding algorithm. Designing and developing the supporting software tools to provide reasonable parameter settings of the workload norms for the enterprise production management research which based on workload control.