Multi-Criteria Decision Making for Maintenance Strategy Selection

Maintenance strategy selection is a critical challenge across sectors from manufacturing and energy to transport and healthcare. Decision makers must balance competing objectives—cost, reliability, safety, sustainability and resource constraints—while accommodating both quantitative data and expert judgements. Multi-Criteria Decision Making (MCDM) provides structured frameworks for evaluating alternative maintenance approaches, including corrective, preventive, condition-based, reliability-centred and predictive maintenance. By decomposing complex problems into hierarchies or networks of criteria, MCDM methods facilitate transparent weighting of stakeholder priorities and systematic comparison of strategies under uncertainty. Recent advances in sensor technologies, data analytics and machine learning have enriched MCDM applications, enabling real-time condition monitoring and failure forecasting to inform optimal maintenance scheduling. Hybrid models that integrate causal analysis, linguistic assessments and outranking techniques have demonstrated enhanced robustness in handling interdependent criteria and ambiguous inputs. Across diverse case studies—from oil refineries and power plants to marine systems—MCDM tools have delivered tangible improvements in equipment availability, lifecycle cost reduction and safety performance, underlining their global significance and adaptability to evolving industrial demands.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent studies have introduced sophisticated hybrid MCDM frameworks to tackle the inherent complexity of maintenance decisions. One approach combines the Decision-Making Trial and Evaluation Laboratory method with the Analytic Network Process and the VIKOR technique under interval type-2 fuzzy sets to address uncertainty and interdependencies among economic, safety and sustainability criteria. In an oil-refinery case study, this model ranked run-to-failure, preventive, condition-based and reliability-centred maintenance strategies for distillation units, revealing that advanced data-driven options consistently outperform conventional choices.

Another contribution integrates artificial neural networks with classic MCDM methods in a large-scale hydroelectric power plant. Here, the Analytical Hierarchy Process weights equipment criticality criteria, Technique for Order of Preference by Similarity to Ideal Solution ranks components by maintenance priority, and a neural model forecasts failure intervals. The resultant plan reduced unplanned shutdowns, extended asset life and optimised workforce allocation over a two-year implementation.

A third model unites fuzzy DEMATEL, fuzzy AHP and fuzzy TOPSIS to establish causal links among criteria, derive priority weights and compute closeness to ideal solutions. Applied to industrial maintenance policy selection, this framework demonstrated improved decision accuracy under linguistic uncertainty and provided decision makers with a transparent sensitivity analysis of maintenance alternatives.

Multi-Criteria Decision Making for Maintenance Strategy Selection publication trend

The graph below shows the total number of articles in multi-criteria decision making for maintenance strategy selection across all publications each year (not limited to Nature Index journals).

Technical terms

Multi-Criteria Decision Making (MCDM): A collection of methods for evaluating multiple conflicting criteria to support structured decision processes.

Analytic Hierarchy Process (AHP): A technique that decomposes a decision problem into a hierarchy, eliciting pairwise comparisons to derive criterion weights.

Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS): An outranking method that ranks alternatives based on their distance from ideal and anti-ideal solutions.

Analytic Network Process (ANP): An extension of AHP that captures interdependencies among criteria via a network structure.

Decision-Making Trial and Evaluation Laboratory (DEMATEL): A method to map and quantify causal relationships among decision criteria.

VIKOR: An MCDM technique focusing on compromise solutions by measuring closeness to the ideal alternative.

References

1. An artificial neural network model supported with multi criteria decision making approaches for maintenance planning in hydroelectric power plants. Eksploatacja i Niezawodnosc - Maintenance and Reliability (2020).