Fuzzy sets and their variants
Irem Ucal Sari; Umut Ak
Abstract
Industry 4.0 implementations are competitive tools of recent production systems in which complex computerized systems are employed. Efficiency of these systems is generally measured by Data Envelopment Analysis (DEA) under certainty. However, the required data in modelling the system involve high degree ...
Read More
Industry 4.0 implementations are competitive tools of recent production systems in which complex computerized systems are employed. Efficiency of these systems is generally measured by Data Envelopment Analysis (DEA) under certainty. However, the required data in modelling the system involve high degree of uncertainty, which necessitates the usage of fuzzy set theory. Fuzzy DEA models can successfully handle this problem and present efficient solutions for Industry 4.0 implementation. In this paper, efficiency of Industry 4.0 applications is measured by classical DEA and fuzzy DEA models, allowing the variables to have different units of measurement and to be independent from analytical production functions. Besides that, fuzzy algorithms for output-oriented DEA are proposed for BBC and CCR models. To the best of our knowledge, this article is the first quantitative academic study to measure the effects of Industry 4.0 applications on productivity. It also shows how fuzzy factors can affect decision-making by comparing fuzzy and classical DEA results. A real application of the models is realized in a company of home appliances manufacturing sector having Industry 4.0 applications. The effect of Industry 4.0 implementation on machine productivity, and superiority of fuzzy DEA over classical DEA are shown through the application.
Fuzzy sets and their variants
Fatemeh Zahra Montazeri
Abstract
One of the appropriate and efficient tools in the field of productivity measurement and evaluation is data envelopment analysis, which is used as a non-parametric method to calculate the efficiency of decision-making units. Today, the use of data envelopment analysis technique is expanding rapidly and ...
Read More
One of the appropriate and efficient tools in the field of productivity measurement and evaluation is data envelopment analysis, which is used as a non-parametric method to calculate the efficiency of decision-making units. Today, the use of data envelopment analysis technique is expanding rapidly and is used in the evaluation of various organizations and industries such as banks, postal service, hospitals, training centers, power plants, refineries, etc.In real-world problems, the values observed from input and output data are often ambiguous and random. To solve this problem, data envelopment analysis in stochastic fuzzy environment was proposed. Although the DEA has many advantages, one of the disadvantages of this method is that the classic DEA does not actually give us a definitive conclusion and does not allow random changes in input and output. In this paper, we review some of the proposed models in data envelopment analysis with fuzzy and random inputs and outputs.
Other
Abbasali Monzeli; Behrouz Daneshian; Gasem Tohidi; Masud Sanei; Shabnam Razavian
Abstract
Data Envelopment Analysis (DEA) is one of the well-known methods for calculating efficiency, determining efficient boundaries and evaluating efficiency that is used in specific input and output conditions. Traditional models of DEA do not try to reduce undesirable outputs and increase undesirable inputs. ...
Read More
Data Envelopment Analysis (DEA) is one of the well-known methods for calculating efficiency, determining efficient boundaries and evaluating efficiency that is used in specific input and output conditions. Traditional models of DEA do not try to reduce undesirable outputs and increase undesirable inputs. Therefore, in this study, in addition to determining the efficiency of Decision-Making Units (DMU) with the presence of some undesirable input and output components, its effect has also been investigated on the efficiency limit. To do this, we first defined the appropriate production possibility set according to the problem assumptions, and then we presented a new method to determine the unfavorable performance of some input and output components in decision-making units. And we determined the impact of unfavorable inputs and outputs on the efficient boundary. We also showed the model result by providing examples for both unfavorable input and output states and solving them and determining the efficiency score and driving them to the efficient boundary by plotting those boundaries.