COST AND SERVICE-LEVEL-BASED MODEL FOR A SERU PRODUCTION SYSTEM FORMATION PROBLEM WITH UNCERTAIN DEMAND

With increasing demand diversification and short product lifecycles, industries now encounter challenges of demand uncertainty. The Japanese seru production system has received increased attention owing to its high efficiency and flexibility. In this paper, the problem of seru production system formation under uncertain demand is researched. A multi-objective optimization model for a seru production system formation problem is developed to minimize the cost and maximize the service level of the system. The purpose of this paper is to formulate a robust production system that can respond efficiently to the stochastic demand. Sample average approximation （SAA） is used to approximate the expected objective of the stochastic programming. The non-dominated sorting genetic algorithm II （NSGA-II） is improved to solve the multi-objective optimization model. Numerical experiments are conducted to test the tradeoffbetween cost and service level, and how the performance of the seru production system varies with the number of product types, mean and deviation of product volume, and skill-level-based cost.

A nested partitioning-based solution method for seru scheduling problem with resource allocation

This paper investigates the production scheduling problems of allocating resources and sequencing jobs in the seru production system(SPS).As a new-type manufacturing mode arising from Japanese production practices,seru production can achieve efficiency,flexibility,and responsiveness simultaneously.The production environment in which a set of jobs must be scheduled over a set of serus according to due date and different execution modes is considered,and a combination optimization model is provided.Motivated by the problem complexity and the characteristics of the proposed seru scheduling model,a nested partitioning method(NPM)is designed as the solution approach.Finally,computational studies are conducted,and the practicability of the proposed seru scheduling model is proven.Moreover,the efficiency of the nested partitioning solution method is demonstrated by the computational results obtained from different scenarios,and the good scalability of the proposed approach is proven via comparative analysis.

Seru Scheduling Problems with Multiple Due-Windows Assignment and Learning Effect

This paper deals with seru scheduling problems with multiple due windows assignment and DeJong’s learning effect.Specific time intervals are assigned to jobs with multiple due windows and learning effect is introduced to characterize the decrease of processing times with the accumulation of the working experience.We assume that the set of jobs assigned to each due window is independent,and no inclusion exists between due windows.The objective is to determine the optimal due window positions and sizes,the set of jobs assigned to each due window,and the optimal schedule in each seru to minimize a multidimensional function,which consists of the earliness and tardiness punishment cost,as well as the due window related starting time and size cost.We find that when the number of jobs and the due windows assigned to each seru are pre-specified in advance,the problem can be solved in polynomial time.Meanwhile,the impacts of the due-window allocation strategy and learning effect on the total cost are respectively discussed based on numerical examples and special cases.The results show that if each seru is assigned with the same number of due windows,the total cost can be reduced with the increasing ratio of the due-window number to the to-be-processed job number.Furthermore,with an increasing learning effect,the total cost will be decreased.