A Novel Integrated Fuzzy-Rough MCDM Model for Evaluation of Companies for Transport of Dangerous Goods. (28th December 2021)

• Record Type: Journal Article
• Authors: Vojinović, Nikolina; Sremac, Siniša; Zlatanović, Dragan
• Other Names: Abramović Borna Academic Editor.
• Is Part Of: Complexity. Volume 2021(2021)
• Publication Date: 2021-12-28
• Subjects: Chaotic behavior in systems -- Periodicals; Complexity (Philosophy) -- Periodicals; 003
• DOI: 10.1155/2021/5141611
• Languages: English
• ISSNs: 1076-2787

Abstract

Abstract : The organization and execution of the transport of dangerous goods is conditioned by a series of legal, technical, technological, safety, and engineering requirements, which must be met. In this way, a complex system is created which has a large number of participants and in which optimization should be performed at each stage from all the above aspects. The main goal of this paper is to create a novel Fuzzy-Rough MCDM (Multiple-Criteria Decision-Making) for the evaluation of companies engaged in the transport of dangerous goods. A group decision-making model was created to evaluate 11 transport companies based on nine legal, technical, technological criteria. The improved fuzzy stepwise weight assessment ratio analysis (IMF SWARA) method was used to calculate the criterion weights, while transport companies were ranked based on Rough Measurement Alternatives and Ranking according to the COmpromise Solution (R-MARCOS). The integration of these methods into a single model that combines two theories of uncertainty, fuzzy and rough, was performed for the first time in this study, which represents a significant contribution. The results show that the most significant criteria are as follows: dangerous goods are classified and permitted under ADR (Agreement Concerning the International Carriage of Dangerous Goods by Road), the prescribed documents are in the transport unit, and the equipment is in the transport unit. When it comes to the evaluation results of companies, it can be noticed that A1 and A4 show the best performance, while A8 and A9 are in the last position. In order to test the stability of the model developed, sensitivity analysis, comparative analysis, and the influence of the dynamic formation of the initial matrix were created.