Research on Multiobjective Group Decision-Making in Condition-Based Maintenance for Transmission and Transformation Equipment Based on D-S Evidence Theory

Abstract

Aiming at eliminating the current drawbacks that exists in condition-based maintenance (CBM) decision-making, such defects as single objective model, difficulty in controlling the process, and individual rather than group decision-making, this paper proposed a multiobjective group decision-making CBM model based on Dempster-Shafer (D-S) evidence theory by establishing maintenance decision-making framework, setting index system and quantifying all the indices. The model takes decision-making indices as evidence sources and maintenance schemes as knowledge framework, solving the problem of difficulty in demarcation of objectives and constraints. Maintenance schemes are achieved according to mean residual life prediction based on maintenance guidance and current condition quantified by health index to estimate the most appropriately maintenance interval. After that, standardized values of indices under each scheme are normalized into mass function of D-S evidence theory. Then, considering comprehensive weights of indices decision-makers, multiobjective group decision-making results of maintenance types, and time intervals are obtained. Finally, a numerical case is studied to illustrate the proposed modeling process, and demonstrates that the model can effectively realize multiobjective group decision-making, as well as considering the intentions of decision makers while all indices can reach global optimum.