Driving net-zero construction through evolutionary machine learning in Vietnam: a strategic framework for sustainable performance

Purpose This study investigates how evolutionary machine learning (EML), a class of adaptive and intelligent optimisation techniques, can be strategically employed to drive sustainable performance in construction firms, particularly in developing economies. Design/methodology/approach Using a business optimisation lens, this research develops and empirically tests a comprehensive framework that integrates ecological modernisation theory, adaptive structuration theory and diffusion of innovation to understand post-adoption impacts of EML-enabled technologies on carbon neutrality and organisational performance. Findings Through structural equation modeling analysis of 213 Vietnamese construction firms, the findings underscore the importance of aligning EML integration strategies with operational realities, stakeholder readiness and long-term innovation capabilities to achieve a net-zero construction supply chain and improved business performance. Research limitations/implications By framing EML as a tool for solving combinatorial and dynamic optimisation challenges, such as resource allocation, project scheduling and carbon footprint reduction, this research contributes to discourse on evolutionary computation for real-world business problems. Originality/value EML-enabled technologies can provide optimal solutions that balance multi-objective problems such as minimising cost and environmental footprint while maximising the economic value of construction that traditional machine learning cannot address. Although prior research on individual technologies and sustainability in construction supply chains has been conducted, there are limited studies centralising applications of EML-enabled optimisation in driving net-zero construction supply chains.