An efficient optimization method for determining the number of stations, the configuration of machines, the number of pallets in the system and the workload allocation was proposed in a generalized separated-functional Flexible Manufacturing System without requirements for balanced loading. Some numerical examples were demonstrated to show that solutions obtained under the requirements for workload balancing are the same as the optimal solutions in most cases and are very close to the optimal solutions in the other cases.
This paper investigates the economic replacement problem from the current equipment to a new one, using the index of internal rate of return, for the case in which periodic maintenance is required whose cost increases time by time while the annual operating cost is constant over the entire period. This problem is comprised of that of obtaining economic life for the new equipment to be used in the manner of like-for-like replacement and that of clarifying economic replacement timing from the current equipment to the new one. Taking into consideration the property that both the economic life and the economic replacement timing can be obtained only at the timing maintenance work is practiced, the paper presents a simple method for obtaining the breaking points of capital interest rate, which can be calculated as an internal rate or return for a cash-flow pattern with initial investment and following periodic return. The optimal replacement policy is directly derived by comparing given capital interest rate and the obtained breaking points.
This paper proposes an analytical model of competitive strategy for late entry. Carpenter et al. (1990) proposed a model analyzing the late entry strategy under preference asymmetry. As their market-share function is abstract, the analytical framework of their model is restricted to monopoly market. Moreover, the lack of dependence between cost and position keeps their model from providing qualitative results for decision making with respect to the positioning and the pricing. In this article, first we extend the framework, by means of multinominal logit model, to oligopoly market. Second, taking the above dependence into consideration, we propose a model that provides quantitative decision making to optimize positioning and pricing for a late entrant into an oligopoly market with preference asymmetry.
The previous paper  proposed an effective method, the BN method (forward BEST "N" method), for variable selections on multiple regression analysis for the purpose of analyzing source of variation factors. This paper obtaines that the BN method has a certain validity by conducting case study. The first, the actual situations in applying multiple regression analysis are investigated. Realistic models are set by the results based on this investigation. Compared between the BN method and the usual variable selection methods by many simulation experiments, the validity of the BN method is shown by the results of this comparison. The figures confirmed by the BN method show better results than the usual methods in the tangled relations between cause and effect. Even in the untangled situations, the results of the BN method get the similar ones obtained by the usual methods. The second, the BN method in applying the practical reports with good results is confirmed that it gains the same results in using real data as well as the simulation experiments. For the reasons mentioned above, the BN method is provided to verify an effective variable selection led logical and exact regression as a further milestone.
In this paper, we propose the Conjoint Model and Anaysus using a product experimental design that allocates consumer-attributes and situational factors in an outer array, product-attributes in an inner array. As a result, its analysis leads to inexpensive examination of interaction between consumer-attributes, situational factors and product-attributes, which can provide valuable insights to suit the product planning for the market segmentation or situational factors. An application of the proposed analysis is also provided in which consumers' preference for a telephone is investigated.
Recently, Arizono et al. have derived the cumulant generating function for the log-likelihood ratio statistics under any alternative hypothesis. Further, they have proposed two approximations for the log-likelihood ratio statistics by using only the first and the second cumulants. As a result, these approximations have the high accuracy under the null hypothesis. In addition, it is known that there is some room for further improvement for accuracy of them under the alternative hypothesis. In this paper, we consider the approximations of the log-likelihood ratio statistics by using the higher order cumulants. Then, we compare some approximations for the log-likelihood ratio statistics.