MADM Framework for Strategic Resource Planning of Electric Utilities

This study presents a multi-attribute decision making (MADM) frameworkin support of strategic resource planning of electric utilities. Study efforts have focusedon four technical issues identified to be essentially important to the process of strategicresource development, i.e., decision data expansion, MADM analysis with impreciseinformation, MADM analysis under uncertainty and screening applications. Maincontributions from this study are summarized as follows. First, an automatic learningmethod is introduced for decision data expansion aiming at reducing the amount ofcomputations involved in the creation of decision database. Test results have shown thatthe proposed method is feasible, easy to implement, and more accurate than thetechniques available in the existing literature. Second, an interval-based MADMmethodology is developed, which extends the traditional utility function model with themeasure of composite utility variance, accounting for individual errors from inaccurateattribute measurements and inconsistent priority judgments. This enhanced decisionapproach would help the decision-maker (DM) gain insight into how the imprecise datamay affect the choice toward the best solution and how a range of acceptable alternativesmay be identified with certain confidence. Third, an integrated MADM framework isdeveloped for multi-attribute planning under uncertainty which combines attractivefeatures of utility function, tradeoff/risk analysis and analytical hierarchy process andthus provides a structured decision analysis platform accommodating both probabilisticevaluation approach and risk evaluation approach. Fourth, the application of screeningmodels is investigated in the context of integrated resource planning of electric utilities asto identify cost effective demand-side options and robust generation expansion planningschemes.