Validity of using expert judgements to inform multiple criteria decision analysis: Selecting technologies for sulfide-enriched fine coal waste reuse

Abstract

Mine wastes are produced in large volumes and carry environmental and social risks. Reuse of these wastes can reduce or remove the associated risks. Mine waste reuse technologies are, however, largely undeveloped. Even where such technologies exist, the heterogeneity of mining and mineral waste materials as well as the associated lack of data must be overcome when selecting the most appropriate technology alternatives for reusing such waste. Expert judgement is an emerging technique for soliciting data to support decision-making, but the reliability of this technique has not been rigorously studied. The objective of this study was to assess the suitability of expert judgement input as data for decision support for selection of mine waste valorisation technologies, using sulfide-enriched coal processing waste as a case study. In this study multiple criteria decision analysis, specifically the value function method, was used and 17 criteria grouped into technical, social, economic, and environmental categories were considered. Experts provided questionnaire-type input to a decision rubric supported by linguistic constructed scales. The data for the overall group of experts were then compared with questionnaire and interview data from a subset of minerals design experts. The suitability of using the elicited expert judgement input to inform selection of preferred mine waste reuse technology options was investigated using distinguishability analysis. Approaches to improving the distinguishability of the expert judgement input included improving participant selection and conducting interviews. This study has indicated that the application of expert judgement to inform MCDA does not provide sufficient certainty for reliable decision-making. Improving the selection of experts as well as conducting interviews improved the distinguishability of the technology alternatives, but the goal of being able to support credible technology alternative selection was not reached. Interview data showed that uncertainty and disagreement among experts were crucial concerns in cases where high variability was found. This indicates that expert judgement input must be used with care in cases of decision making under high uncertainty. Expert interviews did, however, lead to technical insights and provided direction for future technology development. This paper demonstrates that distinguishability analysis is a rigorous method for interrogating uncertainty when soliciting expert judgements for decision-support.