Journal of Operational Research in its Applications
Year:2012 | Volume:8 | Issue:4 (31)|Papers Count:9

Railway transportation has been in special concern during recent years and it is planned to increase the annual average construction of new lines from 150 kilometers during the last ten years to 1000 kilometers in the coming years. The increase of country’s railway lines will increase the need for railway fleets accordingly. Consequently, it is necessary to implement appropriate policies to increase industrial ability for manufacturing locomotives, wagons and rail equipments. Therefore the construction of Railway Rolling Stock Industrial Estate was discussed seriously by the Ministry of Industry and selecting an appropriate location for this purpose put on the agenda of the Ministry. The goal of this article is determination of location of railway rolling stock industrial estate by applying fuzzy group multi criteria decision making methods. Because of the necessity of locating this industrial estate close to the present industrial estate and railway lines in the country, the construction of such industrial estate has assumed in the near area of those industrial estate. Accordingly, the related criteria which affect selecting a place for such a rail industrial estate were studied and extracted through reviewing literature, as well as considering the essential needs of these kinds of estates and also using experts’ opinions. Considering the group decision making method, we consulted with railway experts for making preferences matrixes and then modulated them by using fuzzy numbers according decision makers’ degree of confidence. Then by applying a combined method of group fuzzy analytical hierarchy and group fuzzy TOPSIS, we have ranked alternative locations and validated the results accordingly.

Collect accurate data to describe the behavior of real systems is very difficult. Accurate mathematical modeling, with the increasing complexity of systems is impossible. Therefore, fuzzy computing is the answer to the complex problems of today. Authors have employed fuzzy concepts; chance constrained programming and simulated annealing as appropriate tools for determining the net present worth value of several projects. At the end, the most economical project was chosen. In this article, the investment expenses and annual net financial processes are considered fuzzy taking the credibility definition into consideration. The proposed model is then solved by simulated annealing program and to check its credibility we have compared the results of our model with the results obtained by the Genetic Algorithm as well as the Branch and Bound routine. The results indicate that for small sized problems, Genetic Algorithm has worked relatively with less error amounts and produces better results. It was determined that the effective parameters in the algorithm SA, the final temperature coefficient, β, is more important and if to be sufficiently small, the algorithm can be a very appropriate time to reach a satisfactory conclusion.

In recent years, emerging modern technologies and World Wide Web, it has been a tremendous boom through competition in the banking. Majestic and enormous banks would be following their aims, drastically in order to improve competitive power. In modern banking, efficiency analysis might be a helpful training to manage them. Bank managements are looking for causes of inefficiencies, so that they are going to get rid of. This provides that they can see what the competition’s consequences would be. Eghtesad Novin Bank (ENB), as one of the leading bank in Iran performs well in electronic banking and deploying the banking activities furthermore, goal-directing any kind of its branch, ENB has been attempted to fulfill its claims. The issue of efficiency evaluating in ENB branches during 1386-1387, is going to be the proposition of this thesis. That is, DEA (Data Enveloping Analysis) and NN (Neural Network) as a couple of nonparametric rates for efficiency evaluating have been used. Data regarding 40 branches having been inaugurated in March, 2010, have been accumulated in 2007-2008; plus, the efficiency rate has been evaluated, relating to mentioned branches.

Association rule mining, or in other words, extracting the possiple relationships among the data in a large data base, is one the most important and time consuming tasks in data mining. Many algorithms have been proposed for this task in literature which differ in speed and the quality of solutions. In this Article and attempt is made to propose an evolutionary method based on multi-objective PSO which has high speed and proper quality in generating the rules. The proposed algorithm has been applied and tested on a set of transaction data with four thousand records and the results prove the efficiency of our algorithm.

Preferential voting system is a kind of multi-criteria decision making problem in which experts rank alternatives. Selecting the best alternatives in the group decision making problems is a subject of many recent studies. There are numerous proposed methods which aggregate the opinions of voters and rank alternatives in a voting system. These methods gain some scores for each alternative and rank them based on the obtained scores. In recent years, OR researchers have used Data Envelopment Analysis to rank the alternatives in the Multi Criteria Decision Making (MCDM) environment. This paper uses the concept of Data Envelopment Analysis and fuzzy concepts to introduce a new mathematical method for selecting the best alternative in a group decision making environment. Numerical example shows how the new model can be applied to rank the alternatives and also the privileges of the proposed method compared to similar methods, are studied.

Six Sigma is a data-centric problem-solving methodology and require breakthrough quality improvement and rapid execution occurs most effectively in their environments. So, the amount of resources needed to work on all the defined Six Sigma projects in the hopper far exceeds the resources available. Leadership needs to prioritize and make project selections, to make the best use of available resources. Some efforts have been made regarding the Six Sigma project selection problems, but these efforts explicitly assumed that the criteria in the decision process are independent, which is not true in the real world. In this paper, we propose a new hybrid multiple criteria decision-making (MCDM) model, which addresses the correlation among criteria with the aid of the Decision-Making Trial and Evaluation Laboratory (DEMATEL) method to build a relations-structure among criteria. The Analytical Network Process (ANP) is used to determine the relative weights of each criterion with dependence and feedback. The VIKOR method (VlseKriterijumska Optimizacija I Kompromisno Resenje in Serbian, meaning Multi-criteria Optimization and Compromise Solution) is then used to prioritize the Six Sigma projects. The proposed model can help practitioners to improve their selection process, particularly when criteria are plentiful as well correlated. Throughout the present study, we employ data from a Turkish logistic company to reveal the proposed performance in comparison with priory delivered by Buyukozkan & Ozturkcan.

Different stages of a Decision Making Unit (DMU) can be related to each other in series or parallel structures. These stages, sub processes, are very applicable in real world problems. Not considering these stages and their relations, while performance evaluation is being performed, would lead to wrong assessments which are far away from reality. For performance evaluation of an entity with its sub processes, utilizing those techniques which can be shown progress or regress can be helpful. One of these techniques is Malmquist Productivity Index (MIP). In this article considering DMUS each of which has sub process connected to each other in series, MPI technique has been developed in a way that for each sub process efficiency score and MIP index have been calculated. Moreover for each entity the aggregate efficiency score and the aggregate MIP index have also been introduced and calculated.

In the paper of Pachkova [3] the restricted reallocation of resources, among the independent units of a aggregated unit with the objective of optimizing the whole input consumption and output production of that unit, have been analyzed and studied. In this paper, some reallocation models, similar to models presented in the paper of Pachkova, are proposed. With this difference that these models may present the projecting point of each one of the independent units related to the aggregated unit studied in each region through preserving the stability of efficient frontiers, and indicate the output optimized flow in the reallocation in addition to the input optimized flow. Also, an application of the proposed models on Bank data has been fulfilled.