Scientia Iranica
Volume:28 Issue: 4, Jul-Aug 2021

In some complicated datasets, due to the presence of noisy data points and outliers, cluster validity indices can give conflicting results in determining the optimal number of clusters. This paper presents a new validity index for fuzzy-possibilistic c-means clustering called Fuzzy-Possibilistic)FP (index, which works well in the presence of clusters that vary in shape and density. Moreover, FPCM like most of the clustering algorithms is susceptible to some initial parameters. In this regard, in addition to the number of clusters, FPCM requires a priori selection of the degree of fuzziness (m) and the degree of typicality (η). Therefore, we presented an efficient procedure for determining an optimal value for and . The proposed approach has been evaluated using several synthetic and real-world datasets. Final computational results demonstrate the capabilities and reliability of the proposed approach compared with several well-known fuzzy validity indices in the literature. Furthermore, to clarify the ability of the proposed method in real applications, the proposed method is implemented in microarray gene expression data clustering and medical image segmentation.

In this paper, we expand the Muirhead mean (MM) operator and dual MM (DMM) operator with 2-tuple linguistic Pythagorean fuzzy numbers (2TLPFNs) to propose the 2-tuple linguistic Pythagorean fuzzy MM (2TLPFMM) operator, 2-tuple linguistic Pythagorean fuzzy weighted MM (2TLPFWMM) operator, 2-tuple linguistic Pythagorean fuzzy DMM (2TLPFDMM) operator and 2-tuple linguistic Pythagorean fuzzy weighted DMM (2TLPFNWDMM) operator. Based on the proposed operators and built models, two methods are developed to solve the MADM problems with 2TLPFNs and the validity and advantages of the proposed method are analyzed by comparison with some existing approaches. The method proposed in this paper can effectively handle the MADM problems in which the attribute information is expressed by 2TLPFNs, the attributes’ weights are completely known, and the attributes are interactive. Finally, an example for green supplier selection is used to show the proposed methods.

Response surface approach is effective for robust parameter design. Previous response surface methodology assumes that the independent variables are measured without errors. However, this assumption might be violated due to the low capability of measurement system. This paper is concerned with applying response surface method for robust parameter design when there are measurement errors in variables. We present an unbiased estimator when there are some measurement errors and an optimal setting which is determined to minimize the expected quadratic loss. An example is illustrated to verify the effectiveness of the proposed approach. The results show that the proposed method can achieve better operating conditions under situations with measurement errors than the conventional method.

In the management of suppliers, it is an important task to compare the performance of two suppliers using the linear profiles. In this paper, the product acceptance determination procedure is designed using a EWMA statistic based on the process-yield index applied to the linear profiles of two suppliers. The design parameters of the proposed plan are determined to satisfy both the producer’s and consumer’s risks. The efficiency of the proposed sampling plan is compared with the sampling plan developed based on the Wang’s test statistic in terms of the sample size required for the selection of a better supplier. A real example is given to explain the proposed sampling plan.

This study presents a new hybrid intelligent system with ensemble learning for stock selection using the fundamental information of companies. The system uses the selected financial ratios of each company as the input variables and ranks the candidate stocks. Due to the different characteristics of the companies from different activity sectors, modular system for stock selection may show a better performance in comparison with an individual system. Here, a hybrid soft clustering algorithm is proposed to eliminate the noise and partition the input data set into more homogeneous overlapped subsets. The proposed clustering algorithm benefits from the strengths of the fuzzy, possibilistic and rough clustering to develop a modular system. An individual Takagi-Sugeno-Kang (TSK) system is extracted from each subset using an artificial neural network and genetic algorithm. To integrate the outputs of the individual TSK systems, a new weighted ensemble strategy is proposed. The performance of the proposed system is evaluated among 150 companies listed on Tehran Stock Exchange (TSE) regarding information coefficient, classification accuracy and appreciation in stock price. The experimental results show that the proposed modular TSK system significantly outperforms the single TSK system as well as the other ensemble models using different decomposition and combination strategies.

Origin-Destination Matrix, one of the most important elements in transportation planning, is usually estimated by various techniques such as mathematical modeling, statistical methods, and heuristic approaches. Since using electronic devices is rapidly increased to help decision makers to improve models’ capabilities, an iterative procedure is proposed in this paper to estimate the O-D Matrix according to vehicles’ license plates detection. The main concept is to track vehicles on the first and the last links equipped by plate camera over the shortest path from origins to destinations. A two-step procedure and mathematical models are developed to adjust assigned the passing traffic to the network links by minimizing deviations between the observed and estimated truck traffic volumes. The proposed procedure is explained by an illustrative example followed by validation using experimental road network that covers seven eastern provinces of Iran including 310 nodes, 400 two-way edges, and around 3600 origin and destination pairs. Results revealed that the proposed procedure is capable to estimate O-D matrix when the network links are optimally located and equipped by road camera detection systems. In addition, such as the other heuristic approaches, the proposed procedure is sensitive to the number of iterations on the estimation accuracy.

The portfolio of urban and public projects should be balanced in terms of completion time, districts and strategic objectives. For this purpose, we suggest a mixed integer nonlinear programming model based on the goal programming approach. Projects are selected so as to minimize the squared deviation of urban and regional development indicators from their respective targets. In the proposed model there are two category of indicators: coverage indicators that are measured based on the distance of each neighborhood from the nearest covering facility, and general indicators that are usually measured based on the capacities and capabilities of each district. It is assumed that the location of covering facilities have already been selected, but the construction of these facilities will be prioritized and planned according to budget constraints and in competition with other regional development projects. Numerical results indicate superior performance of proposed genetic algorithm in comparison to GAMS solvers. Finally, the application of the model is illustrated by an example.

The effect of measurement errors on adaptive Shewhart charts have been investigated by several researchers. However, the effect of measurement errors on the performance of variable sample size EWMA control charts has not so far been investigated. In this paper, the performance of the VSS EWMA chart in the presence of measurement errors is investigated using a linear covariate error model and a Markov chain method. It is shown that the performance of the VSS EWMA chart is significantly affected by the presence of measurement errors. The effect of taking multiple measurements for each item in a subgroup on the performance of the VSS EWMA chart is also investigated. Moreover, the performance of the VSS EWMA control chart is compared with several other control charts in the presence of measurement errors. At last, an illustrative example is presented to show the application of the VSS EWMA control chart with measurement errors.

Despite its intrinsic advantageous features as a tool for increasing discrimination power of the basic DEA (data envelopment analysis) model, augmented DEA has two main drawbacks including the presence of unrealistic efficiency scores and the presence of great distance between its efficiency scores and scores obtained by primary model. In this regard, this paper extends a heuristic method for dealing with both issues and improving the power of augmented DEA model in performance evaluation. Since different virtual DMUs lead to different results for ranking, the hierarchical clustering algorithm is applied in this study to select the best virtual DMUs in order to reduce the possibility of having inappropriate efficiency scores. Finally, to demonstrate the superiority of the proposed approach over previous approaches in literature, two numerical examples are provided.

The scheduling problem of periodic services from service providers to customers located in different places and need different services. The service centers are also located in different positions, each of which has limited number of teams with the capability of performing one or some services. The goal is to simultaneously minimize ‘total service costs’ and ‘total earliness/tardiness’ in providing services to customers. Providing an optimal maintenance schedule is a big challenge in those companies with dispersed supply centers. In this paper, a novel bi-objective mixed integer linear programming model along with augmented epsilon constraint method is presented to exactly solve this problem. Then, a bi-objective meta-heuristic technique based on genetic algorithm is proposed and its performance in solving large-scaled problems is assessed. The uncertain parameters are faced through robust possibilistic programming approach to diminish the risk of decision making. Finally, the performance of the proposed model and solution approaches are evaluated through a real case study in maintenance scheduling of CNG stations equipment in Iran.

As a generalization of the intuitionistic fuzzy sets (IFSs), complex IFSs (CIFSs) is a powerful and worthy tool to realize the imprecise information by using complex-valued membership degrees with an extra term, named as phase term. Divergence measure is a valuable tool to determine the degree of discrimination between the two sets. Driven by these fundamental characteristics, it is fascinating to manifest some divergence measures to the CIFSs. In this paper, we explain a method to solve the multi-criteria decision-making (MCDM) problem under CIFS environment. For it, firstly, the divergence measures are introduced between two CIFSs and examined their several properties and relations. Secondly, a novel algorithm is given based on the proposed measures to solve the problems in which weights corresponding to criteria are resolved using maximizing deviation method. Thirdly, a reasonable example is provided to verify the developed approach and to exhibit its practicality and utility with a comparative analysis to show its more manageable and adaptable nature.

This paper develops a new variables repetitive group sampling plan using the exponentially weighted moving average (EWMA) statistic based on the yield index for the submitted lot. The optimal parameters of the proposed plan are determined under three scenarios based on the average sample number. Average sample number (ASN) is minimized to decrease inspection's cost and time by using the optimization problem for the required quality levels and sundry combinations of producer's and consumer's risks. The comparison study is provided to specify the efficiency of the proposed plan. Furthermore, the proposed plan is presented with an example which indicates its applicability in the industry. The proposed plan is compared with the single sampling plan and repetitive group sampling plan based on the yield index. The upshots are tabulated for various quality levels. The obtained outcomes demonstrate that the performance of the proposed sampling plan is more lucrative than the existing sampling plans in terms of ASN.