Fuzzy Optimization and Modeling JournalFuzzy Optimization and Modeling Journal
http://fomj.qaemiau.ac.ir/
Tue, 19 Feb 2019 21:21:03 +0100FeedCreatorFuzzy Optimization and Modeling Journal
http://fomj.qaemiau.ac.ir/
Feed provided by Fuzzy Optimization and Modeling Journal. Click to visit.Fuzzy bi-level linear programming problem using TOPSIS approach
http://fomj.qaemiau.ac.ir/article_543187_116083.html
This paper deals with a class of bi-level linear programming problem (BLPP) with fuzzy data. Fuzzy data are mainly considered to design the real-life BLPP. So we assume that the coefficients and the variables of BLPP are trapezoidal fuzzy numbers and the corresponding BLPP is treated as fuzzy BLPP (FBLPP). Traditional approaches such as vertex enumeration algorithm, Kth-best algorithm, Krush-Kuhn-Tucker (KKT) condition and Penalty function approach for solving BLPP are not only technically inefficient but also lead to a contradiction when the follower’s decision power dominates to the leader’s decision power. Also these methods are needed to solve only crisp BLPP. To overcome the difficulty, we extend Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) in fuzzy environment with the help of ranking function. Fuzzy TOPSIS provides the most appropriate alternative solution based on fuzzy positive ideal solution (FPIS) and fuzzy negative ideal solution (FNIS). An example is included how to apply the discussed concepts of the paper for solving the FBLPP.Thu, 20 Sep 2018 19:30:00 +0100A Hybrid Algorithm for Fault Diagnosis using Fuzzy Clustering Tools
http://fomj.qaemiau.ac.ir/article_545576_116083.html
In this paper, a hybrid algorithm using fuzzy clustering techniques is proposed for developing a robust fault diagnosis platform in industrial systems. The proposed algorithm is applied in a fault diagnosis scheme with online detection of novel faults and automatic learning. The hybrid algorithm identifies the outliers based on data density. Later, the outliers are removed, and the clustering process is performed. To extract the important features and improve the clustering, the maximum-entropy-regularized weighted fuzzy c-means is used. The use of a kernel function allows achieving a greater separability among the classes by reducing the classification errors. Finally, a step is used to optimize the parameters m (regulation factor of the fuzziness of the resulting partition) and (bandwidth, and indicator of the degree of smoothness of the Gaussian kernel function). The proposed hybrid algorithm was validated using the Tennessee Eastman (TE) process benchmark. The results obtained indicate the feasibility of the proposal.Fri, 31 Aug 2018 19:30:00 +0100The Reference Ideal Method and the Pythagorean Fuzzy Numbers
http://fomj.qaemiau.ac.ir/article_545712_116083.html
As it is well known, in spite of having small dimensions, there are daily manysituations that require the solution of a decision-making problem: eating, streetscrossing, assessments, shopping and so on. Generally, the way of working onthese types of problems depends on how the information used to evaluate eachalternative is provided and represented, as for instance is the case with: crispvalues, fuzzy values, Pythagorean values, etc. In this way, different very wellknownmethods have been developed and modified to help to solve this kind ofproblems. Among them, the following may be remarked: AHP, PROMETHEE,ELECTRE, VIKOR, TOPSIS. But there are many other. This paper shows howto apply the so-called Reference Ideal Method (RIM), previously developed bythe authors, when Pythagorean Fuzzy numbers are used to evaluate eachalternative. The paper shows how to solve a decision-making problem throughthe proposed method using such kind of fuzzy numbers and, in order to showhow to practically apply the RIM method, an illustrative example is provided.Sun, 09 Sep 2018 19:30:00 +0100