Weighting risk impact indicators in a food system: expert elicitation and analysis prior to inclusion in a multi-risk assessment

Abstract To cover the farm-to-fork continuum of a food system, it is necessary to tackle a range of risks associated with different domains, such as agriculture, environment, economics, nutrition and food safety. In each domain, several risk impact indicators are required, for instance in the food safety domain, several hazards (leading to a risk) have to be considered: pathogenic bacteria, heavy metals, etc. Obtaining the data to score the indicator is a scientific work that requires data and models. However, that is not enough: to aggregate the risk impact indicators, weights must be assigned to each indicator. In this study, the lentil food system was chosen as case-study, considering that its production and consumption may increase in the future. The expert elicitation process was conducted within a boarder Living Lab experience to address the inherent subjectivity of weighting procedures, while ensuring transparency, inclusiveness, and reflexivity in expert-based decision support. A group of 11 experts were interviewed regarding weights associated with 16 risk impact indicators. Their responses were scored along with their level of knowledge on each indicator, and then a more in-depth analysis was carried out to establish the weights for each indicator. The weights are presented as a probability distribution representing the variability in the 10 expert judgements, adjusted for the uncertainty (more exactly the lack of certainty) in their belief. This method for obtaining and analysing risk impact indicator weights, brings objectivity, transparency and robustness and then brings an added-value to a multi-risk assessment.