An integrated DSS to support a production manager in his decision making regarding capacity planning, lot sizing, and scheduling, is described here. The system was built based on the actDSS decision support approach, which was proposed in order to truely support a user in his decision making process. The system provides a framework to couple numeric calculation and symbolic manipulation effectively. And therefore it is a step forward towards the ideal DSS in which a user and a system cooperatively solve a given decision problem.

View full abstract

Production systems must be always managed by keeping them adjusted to their external environment. Then, in order to adjust themselves to customer's demand change as a main external environmental change, it's required to change production systems. Decision making concerned with alteration of production systems is a difficult and important task for production managers. In this paper, we construct a production system model using production cost as a criterion. Operating this model and considering customer's demand change in volume and price, we offer a supporting tool for investigating system alteration policies relative to equipment cost, labour cost and development cost.

View full abstract

This paper presents a single-echelon, single item, stochastic lead time and deterministic demand (Q, r) inventory model for situations in which, during the stockout period, a fraction β of the demand is backordered and the remaining fraction (1-β) is lost. In this situations, an objective function representing the average annual cost of inventory system is obtained by defining a time-proportional backorder cost and a fixed penalty cost per unit lost. The optimal operating policy variables minimizing the average annual cost are calculated iteratively. At the extreme β=1 and β=0,the presented model reduces to the usual backorder case and the lost sales case respectively.

View full abstract

Recently, the linear programming with interval coefficients is formulated which expresses the fuzziness of the state. Also, an effective method for solving 0-1 linear programming with GUB structure which utilizes its structure is proposed. But the 0-1 linear programming with both GUB structure and interval coefficients is not investigated. In this paper, we show that a nonlinear integer programming problem is transformed to a 0-1 linear programming problem with both GUB structure and interval coefficients, propose an effective method for solving the problem which utilizes the GUB structure and demonstrate the usefulness of this method.

View full abstract

In order to analyze the performance of Pull type production ordering systems against in-process factors, this paper constructs a model of multi-stage production-inventory system with two kinds of in-process factors, production capacity and machine breakdown. And this paper develops two models of Pull type production ordering systems with the difference of information used. Where, this paper considers the feedback of information about the difference between production order and actual production, that was occurred by in-process factors. With simulation experiments for the models, this paper clarifies the influence of in-process factors upon the amplifications of production quantity and inventory level.

View full abstract

This paper with a three-objective, single-stage scheduling problem to minimize the expected value and variance of total flowtime and the maximum probability of tardiness when stochastic operational time of each job follows normal distribution. An efficient algorithm was proposed for generating all of approximate nondominated schedules. Numerical experiments made it clear that this algorithm efficiently generates an approximate set of nondominated schedules with high accuracy.

View full abstract

A closed queueing network model was constructed to model a Flexible Manufacturing System (FMS) consisting of a loading/unloading station, multiserver workstations and a material handling system. The effects of loading/unloading time and transfer time on the system utilization were analysed through this model. A relationship between the mean loading/unloading time and the number of loading/unloading devices, and that between the mean transfer time and the number of transfer vehicles were derived so that the decrease in the system utilization incurred by loading/unloading operation and transfer be smaller than 5%. The range of the number of pallets was also specified so that these relationships hold.

View full abstract

Twofold Look-Ahead Search is proposed as a general-purpose technique based on the human problem solving process, utilizing the superior calculation ability of computers. In the search process, nodes are expanded in a tree using two look-ahead features : a local look-ahead which foresees states in the near future with high accuracy, and a global look-ahead which provides a rough evaluation of goal states. This method is applied to job shop scheduling, and the schedules generated are compared with those of dispatching rules and an optimization method. Results showed the proposed method to have high performance, giving nearly optimal solutions and being much superior to the dispatching rules. In addition, the effectiveness of the twofold look-ahead approach in the search process is also confirmed through comparisons of several variations of the proposed method.

View full abstract