Most of the authors who have written on the subject of inventory control have assumed a sufficient available information for the decision and focused their attention on obtaining the optimal decision. This paper discusses the kind and accuracy of the information for the optimal inventory system. If the decision maker wants the complete information of the inventory system, then he requires the time and cost to obtain the information. Otherwise, if he uses the incomplete information to decide quickly, then he must have the probability that controls a nonoptimum system. So, he wants to minimize a total sum of the cost to have the information and the loss due to incomplete information. This problem is put to the sensibility analysis for the fixed-order-size inventory system and the fixed-order-interval inventory system.

View full abstract

Under the consideration of the difference of work station interfaces, the mixed-model assembly line is classified into two models. For each model, the relation between the cycle time and the maximum work congestion is analyzed by numerical experiments. With reference to experimental results, we consider how to design buffer capacity in the mixed-model assembly line. Tow kinds of buffer capacities, that is, the allowance of the work area and the overtime work are considered. Mixing these buffer capacities adequately, we attempt to assure the reliability of the assembly line economically. Also we show a numerical example.

View full abstract

In this article some new system estimation methods in econometric model are proposed as 1) Systems h-class Estimation Method (SHCE), and 2) Systems G-dimensional h-class Estimation Method (SGHCE). The above two estimation methods are resulted from combining the h-class estimation method with the three-stage least-squares method in conjunction with the relation between the two-stage least-squares method and the h-class estimation method for a single equation. The following two methods are also presented, by which the above mentioned techniques can further be elaborated. 3) Systems Double k-class Estimation Method (SDKCE), and 4) Systems G-dimensional Double k-class Estimation Method (SGDKCE).

View full abstract

This paper deals with a production and inventory system for the agricultural product processing industry. In that industry, the raw materials are yielded mostly seasonally, while the demand for the products is almost constant throughout the year. In this paper, a condition is assumed such that the raw materials have a limited shelf life because of decay and the analysis is made on relationship between the demand rate of products and the processing rate restricted by the size of work force under a certain yield pattern of raw materials. The results of analysis can be used to determine the manpower and initial product inventory level which is the inventory level at the end of the previous year.

View full abstract

It is an effective means to increase the personnel number of group work as one of the ways to meet the needs of the reduction of working term. On the other hand, if the personnel number is too much, the reduction does not come about because of the increase of meeting time. In this paper, first, a classification of the character of work is described from the viewpoint of increase and decrease of total man-hours in case of increase and decrease of the personnel number of group work. Secondly, in the relation between the existence or nonexistence of a team leader and working hours, a criteria on the personnel number for minimum working hours is described in case of learning phenomenon.

View full abstract

In a two-stage lot production system, it is inevitable that idle times arise at either stage because of the differences in set-up times, processing times, or lot sizes of the items produced. This paper presents an analysis of the characteristics of idle time of a two-stage lot production system, and proposes the scheduling procedures to make the acceptance of the special order items possible in such ways as to control the frequency and duration of idle time by the methods of 1) changing lot sizes, 2) shifting the production starting times of standard items produced according to some forecast or by executing both 1) and 2) at the same time.

View full abstract

This presentation is concerned with an evaluation technique of company standardization at the time of settlement and revision of the standards for products and parts. Its objectives are to decide to put the new standards in operation or not, optimize the plan and establish the co-ordination of related departments. It consists of two phases; the stationary applying period and the change-over period of standards. Also, it has two aspects; evaluation of the effects and optimization of the plan. The evaluation procedures and methods for each phase and aspect are explained in details, accompanied with the evaluation method of economical and intangible effects of standardization.

View full abstract

In this paper, an optimal allocation problem of reliability with maintenance is discussed from the viewpoint of the life cycle costing. Concretely, the problem which determines both reliability and preventive replacement time of each equipment is formulated by dynamic programming under the criterion to minimize the life cycle cost from the system. Some properties of the optimal solution and the influence of two replacement policies on the allocation are examined. Numerical examples are also given. The results indicate the importance of introducing the maintenance appropriately into the optimal reliability allocation.

View full abstract

In this paper, the two-echelon production-sales system which deals with growing products is discussed, and the optimal production quantity is determined throughout the model experiment. Assuming the system has a discount sale for the growing products, we discuss the optimization of the tradeoff between the benefit due to a reduction of the inventory and the discounting loss. The available range of discount ratio is determined with this system, and the availability of this system is clarified in comparison with the old production-sales system which ignores the discount sales of the growing products.

View full abstract

This paper gives an experiment of redesign by the joint approach of both job enlargement and simplification in the conveyer line of a firm. We applied the Herzberg's Motivation-Hygiene theory, the contingency theory of behavioral science and the socio-technical systems approach. Motivation seekers adapt to an enlarged job. Hygiene seekers adapt to a simplified job. As a result, the productivity increased and job satisfaction improved. The problem between Industrial Engineering and Behavioral science is not only presenting whether or not to enlarge or to simplify a job, but also to consider the joint approach of both job enlargement and simplification.

View full abstract

Subsampling is an effective statistical method of surveying various phenomena. In this paper, a new theoretical formula has been sought to obtain expectation and variance of means in subsampling with sample size of secondary units uncorresponding to the strata of finite population. This formula can be introduced to various well known existing formulae, special kinds of subsamplings. This method has an advantage in that the total sample size is always the same regardless of the state of sample, but it is disadvantageous in being less precise than the others. In this formula, the optimum allocation using a new cost function proves to make the primary sample size as large as possible, and to of secondary sample of equal size to each other. Therefore, it will become useful in designing a sample survey when parameters are unknown.

View full abstract

In order to ensure the reliability of the mixed-model assembly line, it is very important to analyze properties of the maximum work congestion which occurs from differences of the operation time of each product at each station and variation in demand for each product during the decision period. In this paper, some properties between the maximum work congestion and system factors, that is, cycle time, operation times, and conveyer-in sequences are specified by numerical experiments. From results of the numerical analysis, we know that the maximum work congestion is affected by the critical cycle time and conveyer-in sequences.

View full abstract

This paper deals with a design of single acceptance sampling plans by variables for the doubly exponential characteristics whose standard deviation is known. Some tables are given to enable the suitable sampling plans to be easily designed. By use of the MVLUE (minimum variance linear unbiased estimator) the proposed plans have the following advantages : (1) They are applicable to censored samples as well as complete ones. (2) Sample size is smaller than that for the sampling plan using the sample mean. Numerical examples verify the above merits.

View full abstract

Production ordering systems for multi-stage production processes based on forecasting can be classified into centralized and decentralized types by the pattern of feedback-information flow. The purpose of this paper consists in the theoretical justification for empirical ordering procedure in the two types of production ordering systems, and presentation of a guideline for selection between the two types. Two production ordering rules are derived which minimize each of safety-stock levels in the preceding order of production processes, and the effects of variation in product-demand forecast on safety-stock and production levels are also discussed.

View full abstract

The two stage sample size determination method by C. Stein has been widely utilized to determine sample size. The normality of the parent population is assumed in Stein's method. However, the normality assumption does not always hold good for the practical application. It is known that Stein's method gives unreasonable result for the non-normal population. In this paper, it is discussed how the sample size derived from Stein's method is affected by non-normal population represented by the first four terms of an Edgeworth series, and a table of correction coefficients for Stein's sample size is given. A simple method of the non-normality correction for the sample size is also given by dint of Fisher-Cornish approximation.

View full abstract

A body of data to which factor analysis would be usually applied can be formed in a two-way table cross-classified by the variables and the samples. If another set of variables is added to this, such a multi-mode data can be processed at once only by the Tucker's three-mode factor analysis. This study considers about a three-mode factor analysis and proposes the followings which reduce observed data to factors more effectively : (1) the method to compute factor matrices from which a least-squares solution is obtained and (2) the three-mode factor model different from the Tucker's.

View full abstract

In the prefabricated concrete panel production in the housing industry, both the panels of a special item ordered by a customer and a standard item are produced through the same process, where a metal form is required to produce each panel. The former cannot be produced until the order arrives, while the latter is produced with the forecasting information. In this paper, we analyze experimentally the relations between the parameter indicating the number of available forms for standard item and the distribution of the shop time, the average and the standard deviations of the shop time under some load ratios in a basic model. Therefore the experimental approximate equations for the average and the standard deviations of the shop time are presented.

View full abstract

This paper presents two methods which estimate factor weight in industrial location. One is, by using actual location data, the method for analyzing how each industry has evaluated those factors, and the other refers to an attempt to determine the weights by applying multi-factor channel in information theory. In latter half of this paper, some examples will be shown to study the validity of these two methods.

View full abstract