Most water distribution networks are assemblies of buried elements which makes their diagnosis complex and requires implementing an asset management (AM) policy of actions performed in the short, medium and long terms. To implement such an AM, decision makers must diagnose the condition of the asset by developing specific tools for establishing priorities between assets and plan the actions needed to maintain or improve their condition. This study focuses on two topics. (i) The implementation of a spatiotemporal analysis of the propagation of pipe breaks. This innovative approach inspired by criminality analysis allows focusing on hotspot zones in order to discriminate critical assets. We have designed a new ‘clustering’ criterion that can be used as a prioritization criterion for asset sorting. (ii) The second part addresses the issue of multi-criteria decision making concerning the classification of pipe renewal candidates by comparing two kinds of aggregation method: (i) an unsophisticated approach by using adapted weighted sum methods; and (ii) a sophisticated approach by using the Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) and the ‘Elicitation Et Choix Traduisant la REalité’ (ELECTRE III). The underlying reason for comparing these methods is to know whether the unsophisticated methods ensure reliable replacement policy compared to the sophisticated one.

You do not currently have access to this content.