Integrating AI-driven techniques into building energy consumption forecasting: A pathway toward Net Zero Policy

Purpose: The construction industry is among the highest energy demanding sectors with a global demand of 30% (International Energy Agency, 2024). Such high consumption proves the significance of adopting energy efficiency related strategies or policies, optimization techniques and management to achieve global energy reduction. The UK government released its net zero building policy that all new buildings will be net zero by 2030 and the whole lifecycle of existing structures will follow suit by 2050. To achieve this, £3.9 billion worth of funding has been set aside for decarbonising heat and buildings in a bid to reduce the impact of the built environment on global emissions.
It can be found that the use of AI-driven techniques in energy forecasting is gaining popularity due to their ability to solve complex non-linear problems, which is predominantly seen in the residential and commercial sector. However, limited research exists on leveraging AI-driven techniques to optimize or control policy tools related to energy performance (e.g., building codes, energy performance standards, carbon reduction incentives, and compliance monitoring systems) to drive effective energy use in the construction projects (Awuzie, Ngowi & Aghimien, 2024).
Design/methodology/approach: This study employs a systematic literature review to investigate the impact of AI-driven techniques on policy tools aimed at enhancing building energy performance within UK construction projects. The objectives of this paper include identifying relevant building energy policy tools and exploring how AI-driven techniques can contribute to improved energy performance. To explain the relationship between AI-driven techniques and policy tools, a systems thinking approach will be applied, along with the use of SankeyMATIC to visualize these connections. The analysis will also consider the interplay among variables through the frameworks of AI "technology acceptance model", and how innovation chain model connected to technology acceptance model.
Findings: It can be found that the impact of AI-driven techniques on various policy tools is significantly influenced by actors operating outside the immediate project network, highlighting the importance of considering broader industry dynamics rather than focusing solely on project-specific factors. Furthermore, the relationships between policy tools and countermeasures are complex, characterized by dynamic interactions among stakeholders at multiple levels. This complexity may also imply that the system is capable of evolving in response to changes in policy, technology, and stakeholder behaviour.