Journal of the Operations Research Society of Japan
Online ISSN : 2188-8299
Print ISSN : 0453-4514
ISSN-L : 0453-4514
Volume 37, Issue 2
Displaying 1-11 of 11 articles from this issue
  • Article type: Cover
    1994 Volume 37 Issue 2 Pages Cover4-
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (236K)
  • Article type: Appendix
    1994 Volume 37 Issue 2 Pages App3-
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (248K)
  • Chae Y. Lee
    Article type: Article
    1994 Volume 37 Issue 2 Pages 83-95
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    A single machine n-job scheduling problem is examined to minimize sum of absolute deviations of completion times from a common due date. Simple and hybrid genetic Algorithms are developed by investigating basic operators for the applications of job sequencing problems. For the simple genetic algorithm two heuristic crossover schemes: Algorithm VASX and Algorithm VADX are developed based on important properties of the scheduling problem. Local Improvement techniques are considered to enhance the solution quality of the simple genetic algorithm. The power of a genetic algorithm is illustrated by comparing the performance with branch and bound procedure.
    Download PDF (1071K)
  • Peng Cheng, Shigeru Masuyama
    Article type: Article
    1994 Volume 37 Issue 2 Pages 96-113
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    The problem of computing the expected maximum number Ψ(G,p) of vertex-disjoint s-t paths for a probabilistic graph (G, p) is considered in this paper, where G is a two-terminal graph with specified source vertex s and sink vertex t(s ≠ t) in which each edge has a statistically independent failure probability and each vertex is assumed to be failure-free, and p is a vector of failure probabilities of edges. This computing problem is NP-hard, even though graphs are restricted to several special classes of graphs, e.g., planar graphs, s-t out-in bitrees and s-t complete multi-stage graphs. In this paper, we propose a lower bound Ψ___(G,f,p) of Ψ(G,p) for a probabilistic graph (G,p) based on an s-t path number function f of G. Although the lower bound does not seem to be efficiently computed for a general probabilistic graph, we shall also give a class of probabilistic graphs for which the expected maximum number is efficiently obtained by computing the lower bound.
    Download PDF (1118K)
  • Hiroaki Kawate, Tatsuo Oyama
    Article type: Article
    1994 Volume 37 Issue 2 Pages 114-132
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    This paper applies various types. of space interaction modeling approaches to analyse the relationship between inter-regional telephone call flows and their distances. For the telephone call flows, we consider both frequency and their duration time, and for the distance we define. geographical type and fare type. Applying the gravity model approach to telephone call flows data among 40 regions in the metropolitan area, we find that the distance resistance parameter is larger for the geographical distance case than for the fare distance case, and that the geographical distance gives more effects to the frequency than to the duration time. One-factor type entropy model is found to explain the destination selection behavior well for the origin constrained problem. Furthermore, the inverse entropy model and the constrained entropy model are shown to improve the goodness of fit, and we propose the method to forecast the cost function for applying these entropy models. From these entropy model analyses we know that, the flow depends more on the size of origin rather than the size of destination, and that the distance resistance parameter is larger for the frequency data. Finally this type of modeling approach is shown to be effective for studying and modifying the telephone fare system.
    Download PDF (1575K)
  • Hirokazu Kono
    Article type: Article
    1994 Volume 37 Issue 2 Pages 133-157
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    This paper investigates the lot scheduling problem for multiple products processed on a single machine, under a cyclical production sequence and deterministic demand rate. The problem is formulated in the form of a demand and supply balance equation, with consideration of the method by which each product produced is transported from the single machine's location to spatially different multiple demand locations. The paper identifies conditions that ensure the existence of a feasible production lot schedule, and proves that the "balanced lot size" is a unique form of lot sizes for a feasible schedule over an infinite planning horizon. It is clarified that the balanced lot size is not affected by the transportation method assigned to each product. The difference in transportation method only affects the amount of initial inventory required at the beginning of the horizon. These results are applied to the problem of finding a feasible production schedule over a finite planning period, and two methods, utilization of the balanced lot size and application of the balance equation, are presented and advantages of each method are discussed.
    Download PDF (1485K)
  • Yoshiyasu Yamada, Tomomi Matsui, Manabu Sugiyama
    Article type: Article
    1994 Volume 37 Issue 2 Pages 158-168
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    This paper at first, proposes a new method "Inverted Data Envelopment Analysis(Inverted DEA)" for measuring the relative inefficiency of a set of systems (Decision Making Units). By applying this method, we can identify the relative inefficiency factors of each decision making unit (DMU). The method has, in a sense, an inverse relation with Data. Envelopment Analysis (DEA) developed by Charnes, Cooper and Rhodes. More precisely, our proposed measure of the inefficiency of any DMU is obtained as the maximum of a ratio of weighted inputs to weighted outputs subject to the condition that the similar ratios for every DMU be less than or equal to unity. Secondly, we propose a method to classify DMUs by integrating the information from DEA and Inverted DEA. This method can find the singularity of DMUs. Finally, we show the practicability and effectiveness of the classification method by using real financial and business activity data of the branches of a Japanese department store.
    Download PDF (1011K)
  • Article type: Appendix
    1994 Volume 37 Issue 2 Pages 169-170
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (135K)
  • Article type: Appendix
    1994 Volume 37 Issue 2 Pages App4-
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (272K)
  • Article type: Cover
    1994 Volume 37 Issue 2 Pages Cover5-
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (221K)
  • Article type: Cover
    1994 Volume 37 Issue 2 Pages Cover6-
    Published: 1994
    Released on J-STAGE: June 27, 2017
    JOURNAL FREE ACCESS
    Download PDF (221K)
feedback
Top