AI-driven predictive maintenance using an enhanced TOPSIS approach for complex fuzzy information with Z-numbers

In this paper, we introduce complex fuzzy Z-numbers (CFZNs) and pioneered a novel extension of traditional fuzzy set theory by combining complex fuzzy sets and Z-numbers. We systematically explore the fundamental operations of CFZNs, such as union, complement, intersection, subset, and equality, as well as their operational rules. The averaging prioritized aggregation operators (PAgO), weighted averaging PAgO, geometric PAgO, and weighted geometric PAgO are developed. Theorems and properties, like monotonicity, idempotency, and boundedness of these AgOs, are discussed. The novel distance measures are defined to play a crucial role in modeling, analyzing, and solving complex problems in diverse fields. Moreover, we present an innovative multi-criteria decision-making algorithm founded on CFZNs, harnessing their enhanced uncertainty representation capabilities with reliability. Subsequently, we present a case study on utilizing artificial intelligence (AI). The study highlights the significance of AI in the field of predictive maintenance, demonstrating its capacity to improve the dependability of equipment, reduce downtime, and optimize maintenance practices in industrial environments. Furthermore, the extended CFZN-TOPSIS methodology is proposed under the developed framework. Finally, we conduct a comparison analysis and conclude the whole study.