Exploring the Influence of Household Agent-Based Modelling on Whole-life Decarbonisation of Residential Buildings

Abstract

Besides the operational and embodied carbon associated with the physical building, how significant is the influence of the heterogenous nature of household decision making in whole-life decarbonisation of the housing sector? This paper investigates the effects of explicitly considering these factors through agent-based modelling (ABM) of households integrated in bottom-up building stock modelling considering typologies of physical houses (existing and new/future) and different households, and how these might evolve to 2050 in the State of Victoria, Australia. The state population is represented by household typologies based on socio-economic status, tenure, and decisions as influenced by financial, physical/family needs and behavioural factors. The ABM is implemented using Python MESA software and the different house typologies’ whole-life carbon are calculated using the Australian Zero Emission House (AusZEH) modelling software and the Environmental Performance in Construction (EPiC) LCA database, respectively. Considering household decisions yielded higher operational carbon reduction on average across various scenarios, however these have less impact on reducing embodied carbon. And in Victoria, with increasing trends in population and housing demand, embodied carbon dominates whole-life carbon (WLC) outcomes. The heterogeneity of household decisions cannot be ignored and should be further studied along with embodied carbon reduction strategies and a broad range of scenarios that consider the dynamic and uncertain nature of factors that drive and influence the WLC of buildings.