IEEJ Transactions on Electronics, Information and Systems
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
Volume 114 , Issue 4
Showing 1-15 articles out of 15 articles from the selected issue
  • Toshihide Ibaraki
    1994 Volume 114 Issue 4 Pages 411-419
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Download PDF (2030K)
  • Takahumi Oohori, Hiroaki Yamamoto, Nenso Setsu, Kazuhisa Watanabe
    1994 Volume 114 Issue 4 Pages 420-425
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Hopfield has shown that the combinatorial optimization problem can be solved on an artificial neural network system minimizing the quadratic energy function. One of the difficulties in applying the network to actual problems is that the network converges to local minimum solutions very slowly because the sigmoidal function is used for an input-output function of the neuron. In order to overcome this difficulty, we propose an accelerated Hopfield neural network which can control the speed of convergence near the local minima by an acceleration parameter. Computational results for the combinatorial problems with two and twenty-five variables show that: (1) the proposed model converges to the local minima more quickly than the conventional model, (2) that the acceleration of convergence makes the attraction region of each local minimum change and makes the accuracy of the solution worse, (3) that if an initial point is selected around the center of unit hyper cube, the proposed network converges to a local minimum very quickly with high accuracy, and these good properties keep unchanged by the acceleration parameter.
    Download PDF (1375K)
  • Minoru Tada, Hiroaki Ishii
    1994 Volume 114 Issue 4 Pages 426-429
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    When we consider the assignment problem, the value of efficiency eij from the assignment of person i to job j must be determined by a decision maker. However, when he/she wants to evaluate the efficiency suitably, it may be difficult to give a rigid number. That is, the description of the problem seems to need manifold examinations, e.g., a vocational aptitude test (usually at great cost). Since the efficiency (or personal ability) naturally has a qualitative property, we accordingly consider two assignment problems with satisfaction degrees which are based on fuzzy set theory.
    Download PDF (1166K)
  • Katsuki Fujisawa, Mikio Kubo, Susumu Morito
    1994 Volume 114 Issue 4 Pages 430-437
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    In this paper, we report on an application of tabu search to the graph partitioning problem which has applications on circuit board wiring and program segmentation. We discuss how to adapt tabu search to the graph partitioning problem and compare the performance with simulated annealing, another variant of local search incorporating randomized technique. Numerical experiments show that our algorithm dominates the simulated annealing algorithm in accuracy of solutions and speed on both uniform and geometric instances. In particular, our tabu search implementation works much better than the simulated annealing algorithm on structured (geometric) instances. We also investigate how to tune up our implementation and to optimize the various parameters via extensive numerical experiments.
    Download PDF (1786K)
  • Yang Dai, Kazuo Iwano, Naoki Katoh
    1994 Volume 114 Issue 4 Pages 438-443
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Given a connected undirected multigraph G=(V, E) with positive edge capacities and a positive integer k with k≥2, the minimum k-cut problem is to find a set S⊆E of minimum capacity whose removal leaves k connected components. It is known that this problem is NP-hard. In this paper we propose a new approximate algorithm for solving the minimum k-cut problem by using efficient algorithm for the minimum range k-cut problem. Our method is based on the association of the range value of a k-cut and its cut value when each edge weight is chosen uniformly randomly between 0 and 1. Though the closeness of our approximate solution has not yet been theoretically evaluated, we demonstrate its usefulness by performing computational experiments and comparing with another appoximate algorithm proposed by Saran and Vazirani. Our computational experiments show that the new method produces almost the same solutions or even better solutions for small k like k=3, 4, with running time much less than that for the algorithm of Saran and Vazirani.
    Download PDF (1295K)
  • Yasuko Yoshida, Tomomi Matsui
    1994 Volume 114 Issue 4 Pages 444-449
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The edge coloring problem finds an assignment of colors to the edges of a given graph using as few colors as possible, so that no two adjacent edges receive the same colors. This problem arises in many application settings. Examples are routing in a permutation network, a preemptive scheduling of an open shop, a preemptive scheduling of unrelated parallel processors, and a class-teacher timetable problem. When a given graph is bipartite, strongly polynomial time algorithms for finding an edge coloring have previously been given.
    In this paper, we propose an algorithm for finding all the edge colorings of bipartite graphs without parallel edges. Our algorithm requires O (Km(d*+logn)) time and O (d*m2) space, where n (m) denotes the number of vertices (edges), d* denotes the maximum degree, and K denotes the number of edge colorings. Although the time complexity is proportional to the number of edge colorings, the memory space doesn't depend on-it.
    Download PDF (1550K)
  • Bertrand Simon Huppe, Iwao Okutani
    1994 Volume 114 Issue 4 Pages 450-455
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The purpose of this study is to construct a new algorithm using Genetic Algorithm (GA) which can assist decision makers concerned with resource allocation and manpower scheduling problems.
    There are many methods for solving a variety of manpower scheduling problems. Depending upon the requirements involved in a problem, they can be classified into two basic types; one deals with the minimization of the project duration time with a constraint on the total availability of certain key resources, while the other deals with the minimization of the peak of used resources or “leveling” resource demand with a constraint on the total project duration time. In practice, the requirements are becoming so complicated and marginal that it is sometimes practically impossible to solve the problem by traditional methods. In this paper, we developed a systematic scheduling procedure by use of GA, which is very general in that the procedure can take multiple constrains as well as various forms of goal into account. It is demonstrated in the model testings that the proposed procedure can improve the disadvantages of the existing methods compared.
    Download PDF (1346K)
  • Keiichi Uchimura, Kenichiro Kanki
    1994 Volume 114 Issue 4 Pages 456-462
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    We present a practical algorithm that finds the shortest round trip route rounding arbitrary plural nodes on road network data that we made. This algorithm is based on our intuition. When we look at a road map and decide the shortest round trip routes, we should imagine a frame including all tour nodes. The frame is composed of a convex hull. The time complexity is compared between the proposed algorithm and a greedy method which is one of the approximation methods. The practice time of the above two methods and the Hopfield type neural network which is said to be effective for solving a travelling salesman problem, are compared among them each other, too. The accuracy of the proposed algorithm is then compared to an exhaustive search, the greedy method, and the Hopfield type neural network on road network data. From these results, it is shown that the proposed algorithm is effective for finding round trip routes.
    Download PDF (789K)
  • Hiro ITO
    1994 Volume 114 Issue 4 Pages 463-469
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Connectivity (of node-to-node) is used to examine the robustness of graphs. In some telecommunication networks, some switches are integrated into a logical switching area. In this case, we should examine node-to-area connectivity rather than node-to-node connectivity.
    A new idea, NA (node-to-area)-connectivity, is proposed, and this paper shows that the problem of making a minimum NA-connected spanning subgraph of a given graph belongs to the class NP-hard. It also shows, however, that a near-optimum solution to this problem can be given by a linear-time algorithm. The concept of k-NA-connectivity is also introduced, and this paper shows that any k-NA-connected graph G=(V, E) has a sparse k-NA-connected spanning subgraph G'=(V, E') with |E|=O(k|V|). This is shown by presenting an O(|E|)-time algorithm to find one such subgraph. Although the problem of making a minimum k-NA-connected spanning subgraph of a given graph, of course, belongs to the class NP-hard, this problem can be given an approximate solution by the algorithm.
    Download PDF (1160K)
  • Hiroshi Kise, Jinliang Cheng, Hironori Matsumoto
    1994 Volume 114 Issue 4 Pages 470-475
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The 3-machine flowshop scheduling problem that askes to minimize the makespan is very important in present manufacturing systems because 3-machine flowshops are typical examples of the flexible manufacturing systems. From the point of computational complexity this problem is unary NP-complete, suggesting that it is very hard to exactly solve large scale problem instances that occur in practical situations. In order to develop an efficient branch-and-bound algorithm for this problem we notice some dominance relation whose precondition, although being rare in the practical situation, is a sufficient condition for the optimal schedule, suggesting that if it approximately holds, thus being more realistic, then the dominance relation holds with high possibility. The branch-and-bound algorithm proposed here takes advantage of this possibility that is represented by a membership function in the context of fuzzy inference for obtaining an optimal solution as early as possible in the branch-and-bound search. Extensive numerical experiments show that optimal solutions of problem instances with up to 200 jobs can be obtained with quite high probability.
    Download PDF (1333K)
  • Takeshi Yamada, Bruce E. Rosen, Ryohei Nakano
    1994 Volume 114 Issue 4 Pages 476-482
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The Job Shop Scheduling Problem is one of the most difficult NP-hard combinatorial optimization problems. This research investigates finding near optimal schedules using simulated annealing and a recently developed schedule permutation procedure. New schedules are generated by permuting operations within existing schedules. Simulated annealing (SA) probabilistically chooses one of the new schedules and probabilistically accepts or rejects it, allowing importance sampling search over the job shop schedule space. The initial and final temperatures are adaptively determined a priori, and a reintensification strategy improves the search by resetting the current temperature and state.
    Experimental results show this method, possessing the simplicity and flexibility of SA, can find near optimal schedules for the difficult benchmark problems and can outperform the existing SA adjacent swapping approach as applied to job shop scheduling problems.
    Download PDF (973K)
  • Kiyoshi Yoneda
    1994 Volume 114 Issue 4 Pages 483-490
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    This paper concerns the classic problem of contingency tables with given marginals, also known as the problem of origin-destination table estimation. While traditional methods provide nonnegative real number estimates. the method herein provides integer estimates. The algorithm comprises two stages: the first is the standard iterative scaling which yields real value estimates; the second trims the real estimates into integers by combinatorial optimization. A penalty scheme converts the constrained combinatorial optimization into an unconstrained one so that any standard combinatorial optimization method can be used, such as simulated annealing. Bootstrapping experiments and an analysis of 39 sample sets of 12- and 18-floor elevator traffic measurement indicate that the integer estimate provides a higher resolution than the standard real estimate with roughly the same accuracy.
    Download PDF (1240K)
  • Manabu Ishihara, Jun Shirataki
    1994 Volume 114 Issue 4 Pages 491-496
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    The intelligibility of intermittently eliminated Japanese speech wave is discussed in this paper. In our experiment, a segment of a Japanese speech wave is intermittently removed by varying the erasing duration and the period. Here, ratio of the erasing duration to period are varied from 10:90 to 90:10 every 10 percent. Listening tests for evaluating intermittently eliminated speech wave have been carried out by about 1000 males aged from 21 to 24 years. Experimental results show that the intelligibility falls from 0.1 ms for 1 ms. The intelligibility increase for the erasing period surpassing 2 to 3 ms.
    Download PDF (1465K)
  • Hideki Takeda, Fumihiko Kannari
    1994 Volume 114 Issue 4 Pages 497-503
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    A Ti: Sapphire laser pumped with six-flashlamps was developed. Small signal gain measurement was performed resulting in a peak gain-length product of 1.75. Optical energy available in the laser head was estimated to _??_800mJ, and the laser energy of 300 mJ was extracted using a normal stable resonator. When a grazing-incidence grating resonator with a relatively small incident angle of 82° was applied, we obtained the maximum laser energy of 120 mJ and the linewidth of _??_10pm, which was tunable within the mirror reflectivity bandwidth.
    Download PDF (1394K)
  • Yuji Kamiya, Koji Sibata
    1994 Volume 114 Issue 4 Pages 504-505
    Published: April 20, 1994
    Released: December 19, 2008
    JOURNALS FREE ACCESS
    Download PDF (748K)
feedback
Top