Application of a decision-making framework for multi-objective optimisation of urban heat mitigation strategies

Abstract

The significant number of urban heat mitigation strategies (UHMSs) and varying mitigation performances across different urban settings are challenging for governments to make decisions. In response to this challenge, we have previously developed an artificial intelligence-based decision-making framework that can provide governments with optimal UHMS in their urban contexts. To support and demonstrate the use of the framework, this study developed a prototype implementation and applied the framework in Leppington and Green Square, Sydney, Australia. The results showed that optimised UHMS combinations and key planning and design variables were automatically identified to achieve optimal multi-objective performances in reducing air temperature (0.7–0.9 °C), land surface temperature (7.5–11.3 °C), heat-related mortality rate (5.5–6.4%), thermal discomfort (0–0.5 °C), economic productivity loss (1.8–3.5%), electricity energy bills (5.2–6.1%), and implementation cost (22.2–42.2%). The successful application of the decision-making framework in two urban development cases demonstrates a novel approach for governments to obtain the optimal solution for urban heat mitigation in their cities.