Citation

Nagarajan, Deivanayagampillai and Thangavel, Bhuvaneswari and Suppiah, Yasothei and Anbalagan, Kanchana (2026) Dynamic neutrosophic decision matrix (DNDM) for smart city. In: Data-driven Decision Making and Soft Computing. CRC Press, pp. 156-173. ISBN 978-100363483-6, 978-104106297-4, 978-104106314-8

Text 24.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The Dynamic Neutrosophic Decision Matrix (DNDM) of Smart Cities is a novel decision-making tool to address uncertainty and fuzziness in Multicriteria Decision-Making (MCDM) in urban planning, infrastructure development, sustainability, and governance. Because smart cities possess a dynamic and ever-changing nature, traditional decision matrixes fail to incorporate real-time information, fuzzy information, and conflicting stakeholder opinions. Fuzzy set theory has revolutionized the understanding of systems, the theory of logic, and models for reasoning [1]. In the theory of fuzzy sets, the only truth-functional logical connectives are the minimum (min) and maximum (max) operations. They form the core of the determination of the fuzzy logic functions and the consistency of the decision-making. A simple proof shows that distributivity, monotonicity, and boundary conditions are the crucial assumptions in the preservation of the logical structure for fuzzy systems. They ensure the fuzzy operations are well-defined, preserving the crucial mathematical relations necessary for proper reasoning and computation. Fuzzy logic systems can effectively model real-world applications for uncertainty and imprecision by adhering to these principles [2]. This chapter suggests a rules-based approach whereby rules with the same functions are grouped into modules to ensure computational efficiency and transparency in the knowledge base [3]. A fuzzy set for representing the concept is identified by the distribution function for the degree of belief in the qualitative parameter along the interval variable for the quantitative parameter. Since scaling for the concept can be carried out with many parameters, the shape of the fuzzy set is scale-dependent. In environmental research, various stakeholders like the authorities for health, epidemiologists, regulators, politicians, environmentalists, engineers, and the general public describe concepts like contamination in different terms. There is only one generally accepted and well-defined concept in this context, and that is “risk.” Precise scaling and definition for the fuzzy sets provide the common framework for consensus so that the same terms can be used by the experts with different backgrounds to communicate and analyze the risk uniformly throughout the risk assessment process. Fuzziness for the input values is preserved by the fuzzy method with the output in the form of the fuzzy system, which can be transformed into either a quantitative value for the risk or qualitative linguistic terms [4]. This chapter presents two extensions to our Michigan-style Fuzzy Genetic-Based Machine Learning (FGBML). First, the extension is to the heuristic generation of rules. In the typical FGBML, each initial population rule is heuristically generated with respect to an arbitrarily selected training pattern. Additionally, throughout the evolutionary process, the use of heuristic generation is also used, with the rules being obtained from misclassified patterns. Our suggested extension introduces the utilization of the stronger method by generating each fuzzy if-then rule with respect to multiple patterns rather than one pattern [5]. This work also examines the utilization of Genetic Algorithms to optimize Fuzzy Inference Systems for greater accuracy, efficiency, and decision-making. Since Fuzzy Inference Systems are generally used to handle imprecise and uncertain information, Fuzzy Inference Systems can be greatly benefited by the adaptive capabilities of Genetic Algorithms. With the simulation of natural selection, Genetic Algorithms efficiently search complex multimodal problem spaces to find the optimal solution [6]. In this chapter, we introduce the neutrosophic Bonferroni operator for solving problems in multi-attribute decision-making. We define its arithmetic operations, explore its existence and workability, and investigate its most crucial properties and special cases. We utilize arithmetic ranking operations in the neutrosophic framework to address the problems of multiple-attribute decision-making and compare it with other approaches. Our method enhances decision-making by providing a more accurate and reliable optimal solution than the conventional systems.

Item Type:	Book Section
Uncontrolled Keywords:	Computational efficiency, Decision making
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Engineering and Technology (FET)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	03 Sep 2026 07:07
Last Modified:	03 Sep 2026 07:07
URII:	http://shdl.mmu.edu.my/id/eprint/16634

Downloads

Edit (login required)