Journal of the Operations Research Society of Japan

[title in Japanese]

Article type: Cover
1996 Volume 39 Issue 2 Pages Cover4-
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.Cover4

JOURNAL FREE ACCESS

Download PDF (143K)
[title in Japanese]

Article type: Appendix
1996 Volume 39 Issue 2 Pages App3-
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.App3

JOURNAL FREE ACCESS

Download PDF (154K)
CONSTRUCTING CACTUS REPRESENTATION FOR ALL MINIMUM CUTS IN AN UNDIRECTED NETWORK

Hiroshi Nagamochi, Tiko Kameda

Article type: Article
1996 Volume 39 Issue 2 Pages 135-158
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.135

JOURNAL FREE ACCESS

Show abstractHide abstract

The construction of a cactus representation for all minimum cuts in an edge-weighted, undirected network is studied in this paper. Given such a network with a set V of vertices and a set E of edges, the previously known algorithms for constructing a cactus representation require solving |V| - 1 maximum flow problems, each with a prescribed pair of source and sink, in order to find all minimum cuts that separate them. We first show that computing a maximum flow between a pair of vertices, s and t, reveals many other minimum cuts that do not separate s and t. Based on this fact, we propose a new algorithm that can choose an arbitrary pair of source and sink for each maximum flow problem. For a network with unit edge weights, our algorithm runs in O(|E| + λ|V|^2) time, where λ is the weight of a minimum cut.

View full abstract

Download PDF (1873K)
Group DEA for Building Consensus by DMU Efficiency Evaluation

Manabu Sugiyama, Yoshiyasu Yamada

Article type: Article
1996 Volume 39 Issue 2 Pages 159-175
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.159

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper proposes a new method for building group consensus focusing on evaluating systems with the observed values of the multiple inputs and outputs. Here, the systems means Decision Making Units(DMUs) in Data Envelopment Analysis(DEA). This method is composed of two parts. In the first part, it evaluates the relative efficiencies of DMUs by group members with DEA. Next, by using weights of the DEA solutions, we define and calculate the mutual evaluation information of group members. In the second part, we give the method for making a group evaluation which is expressed as the group efficiencies of DMUs by accommodating the difference of their evaluates. It is widely known that the weights are not always uniquely determined. The mutual evaluation information which, we introduced, are calculated from the weights are not always determined uniquely neither. Here, we proposed a method for determining the weights uniquely by minimizing square of the weights differences. We express the mutual evaluation information as a form of matrix which we call it as "Accommodation Efficiency Matrix." Furthermore, this paper discusses how to calculate the group efficiencies of DMUs by using maximum eigenvalue of this matrix. At the end, this paper shows an example which is the productivity analysis of Japanese electric power industries by applying this method.

View full abstract

Download PDF (1604K)
AN ALGORITHM FOR A MULTIOBJECTIVE, MULTILEVEL LINEAR PROGRAMMING

Zhi-Wei Wang, Hiroyuki Nagasawa, Noriyuki Nishiyama

Article type: Article
1996 Volume 39 Issue 2 Pages 176-187
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.176

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper discusses a solution method to a multiobjective, multi-level decentralized system. A decomposition method is proposed for dividing the n-level problem finally into n - 2 single level problems and a two-level problem. A method with domination trees based on the "two-level simplex algorithm" is developed for generating the nondominated solutions to this type of n-level decentralized system.

View full abstract

Download PDF (763K)
MAP/G/1 QUEUES UNDER N-POLICY WITH AND WITHOUT VACATIONS

Shoji Kasahara, Tetsuya Takine, Yutaka Takahashi, Toshiharu Hasegawa

Article type: Article
1996 Volume 39 Issue 2 Pages 188-212
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.188

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper considers MAP/G/1 queueing systems under the following two situations: (1) At the end of a busy period, the server is turned off and inspects the queue length every time a customer arrives. When the queue length reaches a pre-specified value N, the server turns on and serves customers continuously until the system becomes empty. (2) At the end of a busy period, the server takes a sequence of vacations. At the end of each vacation, the server inspects the queue length. If the queue length is greater than or equal to a pre-specified value N at this time, the server begins to serve customers continuously until the system becomes empty. For each case, we analyze the stationary queue length and the actual waiting time distributions, and derive the recursive formulas to compute the moments of these distributions. Furthermore, we provide a numerical algorithm to obtain the mass function of the stationary queue length. The numerical examples show that in light traffic, correlation in arrivals leads to a smaller mean waiting time.

View full abstract

Download PDF (1439K)
PARTIAL PROXIMAL METHOD OF MULTIPLIERS FOR CONVEX PROGRAMMING PROBLEMS

Satoru Ibaraki, Masao Fukushima

Article type: Article
1996 Volume 39 Issue 2 Pages 213-229
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.213

JOURNAL FREE ACCESS

Show abstractHide abstract

Two variants of the partial proximal method of multipliers are proposed for solving convex programming problems with linear constraints, where the objective function is expressed as the sum of two convex functions. The iteration of each algorithm consists of computing an approximate saddle point of the argumented Lagrangian. The global convergence is established under an approximation criterion for computing the saddle point. In particular, for the convex programming problem with multiple set constraints and the traffic assignment problem, one of the proposed algorithms can effectively be implemented on a parallel computer.

View full abstract

Download PDF (1005K)
GRAPHICAL/STATISTICAL APPROACH TO REPAIR LIMIT REPLACEMENT PROBLEM

Hirofumi Koshimae, Tadashi Dohi, Naoto Kaio, Shunji Osaki

Article type: Article
1996 Volume 39 Issue 2 Pages 230-246
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.230

JOURNAL FREE ACCESS

Show abstractHide abstract

The total time on test (TTT) concept has proved to be a very useful tool in many reliability applications. The graphical solution procedure based on this concept gives geometrical interpretations to the solutions of reliability problems as well as the optimal policies from empirical data directly. In fact, some kinds of reliability models such as age replacement and burn-in problems have been analyzed by applying the TTT concept. In this paper, we discuss the optimal repair limit replacement policy, which determines the timing to stop repairing a unit after it fails. We propose two kinds of solution procedures for the repair limit replacement problem with a time constraint, where one is a straightforward/algebraic procedure and another is a geometrical one based on the TTT concept. Also, we develop a computer software utilizing the computer graphics to calculate the optimal repair limit policy numerically, and refer to its educational effect.

View full abstract

Download PDF (891K)
An Approach Based on Necessity Measure to the Fuzzy Spanning Tree Problems

Takeshi Itoh, Hiroaki Ishii

Article type: Article
1996 Volume 39 Issue 2 Pages 247-257
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.247

JOURNAL FREE ACCESS

Show abstractHide abstract

On the decision problems with defective information, the uncertain elements are often formulated as the random variables (the stochastic programming problems) even if they don't behave themselves stochastically. However, when uncertainty is mainly derived from the lack of amount of information and so on, we think, it is proper to recognize them as a kind of "fuzziness" and formulate them as the possibility variables on the fuzzy theory. In this paper, we propose the fuzzy spanning tree problems, in which the edge costs of the graph are possibility variables. At the time of its formulation, we pay attention to the analogy between the random variables and the possibility variables and adopt a model corresponding to the probability maximum one on the chance constrained programming problems for the stochastic programming, that is, the necessity measure maximum model on the modality constrained programming problems for the possibilistic programming. Moreover, in the solution, we propose the efficient algorithm based on the binary search, which fully exploits its problem structure.

View full abstract

Download PDF (690K)
APPLICATION OF THE GRAPH COLORING ALGORITHM TO THE FREQUENCY ASSIGNMENT PROBLEM

Taehoon Park, Chae Y. Lee

Article type: Article
1996 Volume 39 Issue 2 Pages 258-265
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.258

JOURNAL FREE ACCESS

Show abstractHide abstract

The frequency assignment problem is introduced and solved with efficient heuristics. The problem is to assign channels to transmitters using the smallest span of frequency band while satisfying the requested communication quality. A solution procedure which is based on Kernighan-Lin's two way uniform partitioning procedure is developed for the k-coloring problem. The k-coloring algorithm is modified to solve the frequency assignment problem. The performance of the proposed algorithm is tested with randomly generated graphs with different number of nodes, density types and graph types. The computational result shows that the proposed algorithm gives far better solution than a well-known heuristic procedure.

View full abstract

Download PDF (670K)
A NONPREEMPTIVE PRIORITY MAP/G/1 QUEUE WITH TWO CLASSES OF CUSTOMERS

Tetsuya Takine

Article type: Article
1996 Volume 39 Issue 2 Pages 266-290
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.266

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper considers a, nonpreemptive priority queue with two classes of customers. Customers in each priority class arrive to the system according to a Markovian arrival process (MAP). Since the MAP is weakly dense in the class of stationary point processes, it is a fairly general arrival process. The service times of customers in each priority class are independent and identically distributed according to a general distribution function which may differ among two priority classes. Using both the generating function technique and the matrix analytic method, we derive various formulas for the queue length and waiting time distributions. We also discuss the algorithmic implementation of the analytical results along with numerical examples.

View full abstract

Download PDF (1589K)
[title in Japanese]

Article type: Appendix
1996 Volume 39 Issue 2 Pages 291-294
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.291

JOURNAL FREE ACCESS

Download PDF (306K)
[title in Japanese]

Article type: Appendix
1996 Volume 39 Issue 2 Pages App4-
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.App4

JOURNAL FREE ACCESS

Download PDF (180K)
[title in Japanese]

Article type: Cover
1996 Volume 39 Issue 2 Pages Cover5-
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.Cover5

JOURNAL FREE ACCESS

Download PDF (135K)
[title in Japanese]

Article type: Cover
1996 Volume 39 Issue 2 Pages Cover6-
Published: 1996
Released on J-STAGE: June 27, 2017

DOIhttps://doi.org/10.15807/jorsj.39.Cover6

JOURNAL FREE ACCESS

Download PDF (135K)

Register with J-STAGE for free!