The rapidly developing global market place has made the industrial environment more competitive. As a result, the last decade has seen a growing interest in logistics and supply chain management in order to reduce cost and improve responsiveness to uncertain customer demand across the supply chain. The flow of business, such as information and logistics, is improved from the viewpoint of the whole supply chain also including the customer, and it becomes a target to perform optimal response as a whole system. Inventory policy, especially, is crucial to all retailers and suppliers. Recent research trends in inventory management and supply chain management have also focused on the impact of customer behavior on the system or chain of interest. In a recent paper, Lee & Whang identified four common sources of information distortion that lead to the well-known "bullwhip effect". On the other hand, Cheung & Zhang focused on another source of demand information distortion : customer order cancellation. In their research, they assume that each unit of cancellation will cost the retailer as an administrative fee. However, it is general to impose a cancellation cost on the customer in such cases that the retailer has suffered heavy damage from customer order cancellations. From this viewpoint, in this study we discuss the effect on (s, S)inventory policy from customer order cancellations under the assumption that the cancellation cost will be incurred by the customer.

View full abstract

Mold-dies are important tools for producing various types of press and injection products. They generally consist of dozens of parts, each of which is machined in various fabrication stages, and are assembled through many subassembly stages. So, in order to manufacture them efficiently it is important to complete assembly parts nearly at the same time just before sub-assemblies start. Furthermore, in order to increase orders from customers and capture the customer's confidence it is necessary to manufacture mold-dies in a short time and reduce tardy mold-dies. In this paper, we study a scheduling method for a manufacturing system which manufactures various mold-dies requiring subassemblies under dynamic manufacturing conditions. First, we consider three types of dispatching rules which can be applied to the system and may reduce assembly time losses, and present the rule which minimizes the mean of make span using an evaluation scale α; α is the ratio of a make span to an ideal minimum make span. Next, we propose a new method to estimate the minimum due date coefficient which yields an Lp smaller than a given value; Lp is the ratio of the number of tardy mold-dies to the number of completed mold-dies. Finally, we present the usefulness of the proposed method and effects of due date coefficient and load on mold-die tardiness. The main results are as follows : (1) MMO dispatching rules, which use the mold due date revised according to the tardiness of machining stage, minimize the mean of make span. (2) The proposed method can approximately estimate the minimum due date coefficient which yields an Lp smaller than a given value. (3) MMO dispatching rules minimize Lp and the mean tardiness, and MO and DDATE dispatching rules, which use mold and stage due dates calculated from a given due date, minimize the maximum tardiness.

View full abstract

In this paper, we define fuzzy interval data and introduce an approximate method for Bayesian selection of independent variables in linear regression models by using Zadeh's probability concept of fuzzy events. It is important which combination of independent variables is best when we apply the regression analysis. In this paper, we call the data whose boundaries are vague the fuzzy interval data. Here, dependent variables are observed as fuzzy interval data, so we show the posterior distribution for a combination of independent variables using the conventional independent variables and fuzzy dependent variables. This distribution is useful for selection of independent variables. But, the method with direct usage of the membership functions of fuzzy interval data is insufficient from a viewpoint of calculation. However, our proposed method's treatment of the middle points of membership functions as the representative points can solve such a problem. As a result, even if we obtain fuzzy dependent variables, we can formulate an approximate selection of independent variables which is not so far different from the conventional Bayesian selection of independent variables. In realistic situations, we do not always treat ideal symmetrical membership functions. Therefore, we carry out the computer simulations under realistic situations which do not satisfy completely the condition of the symmetry of trapezoidal membership functions. Consequently, we can show the practicability of our method.

View full abstract

The authors presented in their previous paper a method of enumerating alternative ways of performing an assembly-type work, being given a series of work steps together with the appropriate information on all parts and equipment involved in it. The basic idea is characterized by the identification of a set of "essential transformations (ETs)" in a given task. There are numerous ways to build a product unit utilizing a set of ETs depending upon the combination of the sequence of ETs. This paper purposes an algorithm for listing alternative methods to be executed on a computer system and presents a procedure for applying the system in designing and/or improving work processes. As a basic idea in achieving the above, a "state/change transition matrix" is constructed which is processed by the algorithm. By the use of the system, when a set of ETs with a proper set of sequences is input, the system will output a set of corresponding alternative work sequences. For example, if a set of ETs and a set of feasible sequence alternatives are extracted from an existing method of work, and they are input into the system, the system will output the corresponding number of alternative methods of work.

View full abstract

The authors have developed and published a method of converting the information on an existing work method into information on the exhaustive alternative methods for relatively simple hand assembly work. Further, this method enables them to enumerate all of the theoretically efficient and feasible work methods starting from the above information. The authors were searching for such a system that would enable them to visualize each alternative assembly process in 3-dimensional space and were successful in locating one. The system is called the Delimia simulation system and it can reproduce the human work process in a virtual factory-like manner when given specific work points and nature information. However, this system lacks the vital function of automatically generating work method information. Thus, there has been a considerable modification of the method's output information in order to fully utilize the system. The modification itself produced some additional utility of the method which is reported in this paper.

View full abstract

This paper deals with the problem of determine in the number of set-ups for load ratio and set up time ratio, production lot size, production priority and production starting time under a single stage lot production system, multi-production planning period, cyclic schedule for items with a succession of different demand periods. The product items produced in the system have different demand periods and starting times for demand. The production planning aims to determine the number of set-ups for load ratio, set up a time ratio and maintain each demand point for item. As a result of analysis, a structural formulation of product inventory is revealed, as well as the occurrence of interference situations in each production planning. It can clarify the relation among the number of set-ups for each product, production lot size for each product, required quantity of each product, starting time for demand of each product, processing time and set up time of each product, each demand period, product inventory, and two dissolve methods. Based on the formulation, an algorithm is proposed to determine the number of set-ups for load ratio and set up time ratio, production lot sizes, production priority and production starting times. A numerical example shows the usefulness of the procedure proposed.

View full abstract

Production skills can rise to the expert level of knack and intuition only with the right training. More highly skilled workers help contrive and improve the work methods and elucidate customers' order. Experts are also able to train younger workers in every point of work. Knack and intuition for production are also obtained with long-time work experience through the recognition of work essence. Here, a methodology for the construction of a training system for the transfer of skills to a younger generation through the use of cognitive science is discussed. This situation can be same in the development of athletic skills, which has a stronger relationship to intuitive body movement. We compare the two skill acquisition processes from a cognitive scientific point of view. The skill acquisition process includes the so called tacit knowledge of knack, so that transfer by word and manual can have only limited value. OJT is said to be efficient for transfer, but it takes a longer time, as the process and methodology of training has still not been clarified sufficiently. In this study, we propose : (1) a methodology for skill identification using QFD and (2) a transfer strategy using cognitive science.

View full abstract

Suppose that a manager controls a job-shop production system with the backlog by order-selection and switch-over. This paper discusses the 2-stage design problem of the job-shop production model with backlog control from the views of theory and an application study. For this purpose, we apply the 2-stage design method (Matsui, 1999) to the two models : periodic and semi-dynamic types. By the 2-stage design method, the economic traffic that maximizes the net reward (=reward-cost) is decided at the first stage. At the second stage, the economic leadtime is set under the economic traffic. First, a behavioral comparison of discrete and continuous types is given by simulation, and the 2-stage design method for continuous type is showed. Next, the periodic and semi-dynamic models of A company are given under three cases and eight arrival patterns, and the usefulness and robustness of the 2-stage design method are exhibited.

View full abstract