A hierarchical elicitation process based on the combination of different weighting methods within multi-attribute utility theory in a group decision-making context
针对多属性群体决策中信息不完全、决策者专长各异的问题，提出一种分层启发过程，允许决策者仅对自己专长的目标赋权并选择合适方法，通过实际案例（放射性核素污染水体修复）和蒙特卡洛模拟验证了有效性。
Abstract In this paper, we address a group decision-making context within multi-attribute utility/value theory (MAUT/MAVT) under partial or incomplete information. The problem involves objectives of markedly different natures and multiple decision-makers (DMs), each possessing expertise in specific subsets of objectives. We propose a hierarchical elicitation process that integrates multiple weighting methods. Within this framework, DMs contribute exclusively to the elicitation of the objectives for which they hold expertise and may choose the weighting method that best aligns with the type and amount of information they are able or willing to provide. This design further reduces the cognitive burden on DMs, particularly given that each elicitation step focuses on a limited set of objectives–typically of similar nature–within the relevant level and branch of the objective hierarchy, rather than across the entire criteria set. The proposed hierarchical weighting approach has been implemented in a web-based decision support system (WEB-MAUT-DSS) and is illustrated through a real-world decision-making problem: the restoration of aquatic ecosystems contaminated by radionuclides. Monte Carlo simulation techniques are employed to assess its performance across a range of scenarios.