Multi-criteria decision analysis for hybrid water supply systems

Abstract

Traditional urban water systems based on centralised only water supply are facing challenges to meet the increasing water demand due to increasing population in urban centres. These systems are also vulnerable to periodic drought and flooding due to climate change. For this reason, urban water managers around the world have been supporting the adoption of decentralised water supply options such as wastewater reuse, greywater reuse, rainwater harvesting and stormwater harvesting in combination with centralised system to aid with meeting water demand as well as flood mitigation and stream health restoration. In this paper, the combination of centralised water supply system (WSS) with the decentralised system is termed as hybrid water systems. These systems are relatively new and need to be evaluated using a comprehensive framework that can account for multiple aspects of hybrid water supply systems including interaction between centralised and decentralised system. Integration of multiple objectives (such as supplying fit for purpose water, supply reliability of fit for purpose water, wastewater and stormwater discharge and contaminant loads in wastewater and stormwater) to evaluate the hybrid water supply systems can be effectively accomplished by multi-criteria decision aid techniques. Multi-criteria decision analysis (MCDA) requires users to assign the weights to considered criteria to reflect their relative importance. This paper describes weight elicitations in evaluating the various combinations of decentralised and centralised water supply systems. Six performance measures (PMs) were identified in consultation with Victorian water utilities to evaluate the system performance. The identified PMs are: i) Reduction in potable water demand from centralised water supply system, ii) Reduction in wastewater discharges both flow rate and volume, iii) Reduction in contaminant concentration of wastewater flow, iv) Reduction in stormwater flows both intensity and volume, v) Reduction in contaminant loads from stormwater and vi) Improvement of supply reliability of fit for purpose water. Then, a questionnaire survey was conducted among three major stakeholder groups namely, water resource managers, water professionals, and consultants to derive the weights of the selected performance measures. A simple usual preference function was deemed most suitable by the surveyors in order to assess various aspects of hybrid water supply scenarios. Also, it was found that preference measures became more complicated requiring knowledge from diverse disciplines. To overcome this challenge, the use of simple usual preference function was most suitable. Further, the survey results provided almost similar weights to each criterion, varying between 0.15 and 0.18. This indicates the similar importance of all criteria. Also, none of the criteria was deemed as unimportant demonstrating the robustness of all the selected criteria. The elicited weights and preference function of the performance measures in this paper, provide necessary data required to model and evaluate the interactions between hybrid water supply systems.