2026 ASEE Annual Conference & Exposition

Artificial Intelligence (AI) is transforming the Architecture, Engineering, and Construction (AEC) industry, particularly through integration with Building Information Modeling (BIM) to enable data-driven sustainability. However, AI-BIM implementation in emerging economies like Indonesia remains fragmented and constrained. As Asia's second-largest construction market, Indonesia faces the dual challenge of accelerating digital transformation while overcoming socio-technical barriers that limit equitable access to AI-driven technologies. Current AI-BIM frameworks are typically computationally intensive, hardware-dependent, and poorly aligned with local design practices, restricting their scalability in medium-scale public projects where resource efficiency and affordability are essential. This research addresses this gap by developing AI-BIM systems that are contextually responsive, lightweight, and capable of operating within infrastructural constraints while delivering measurable sustainability outcomes.

This study proposes a novel Small Language Model–Building Information Modeling (SLM-BIM) framework that redefines how intelligent systems can support sustainable construction in Indonesia's public sector. Unlike conventional Large Language Model (LLM)-based workflows that require extensive computational resources, SLMs are compact, adaptive, and capable of generating real-time sustainability assessments within modest computing environments. The primary objective of this research is threefold: (1) to quantify performance improvements in terms of energy efficiency (kWh/m²), material waste reduction (kg/m²), and carbon emission reduction (kgCO₂e/m²) compared with baseline BIM workflows; (2) to develop a scalable and replicable AI-supported decision-making model for sustainable design and construction; and (3) to evaluate the regulatory, institutional, and workforce readiness necessary to integrate SLM-BIM at the national level.

The study employs a mixed-method research design, integrating a systematic literature review, case study analysis of selected medium-scale public buildings in Indonesia's new capital region (IKN Nusantara), and empirical simulations of material, energy, and emission performance using BIM-based digital twins. This triangulated approach enables the validation of quantitative and qualitative sustainability performance indicators while ensuring methodological robustness across spatial, temporal, and operational dimensions of construction performance assessment. The empirical findings show that the SLM-BIM framework can reduce computational costs, increase material efficiency, and lower operational carbon emissions relative to conventional BIM methods. Beyond technical outcomes, the research also highlights institutional implications: the integration of SLM-BIM enhances data transparency, supports real-time environmental decision-making, and promotes workforce upskilling through AI-assisted design tools that are accessible to even mid-tier construction firms thereby bridging the current technological divide between advanced and emerging economies in digital construction practices.

The preliminary findings lay the groundwork for policy recommendations, such as establishing an AI-sustainability protocol for public buildings, offering incentives for digital capacity building, and providing regulatory support for integrating SLM-BIM into Indonesia's green construction agenda. The findings also highlight the importance of digital literacy, human-AI interaction, and critical thinking as essential competencies for construction education and workforce readiness. The framework contributes to the body of knowledge by adapting and integrating existing models into a transferable, task-level method and to educational practice by offering evidence-based guidance for curriculum design, industry training, and policy development. The study also addresses the limitations of the proposed method and identifies areas for future research. The research supports Indonesia's long-term climate mitigation strategy by situating this work within the global discourse on sustainable and equitable AI deployment. Also, it advances the development of scalable, resource-efficient AI frameworks for sustainable construction in other emerging economies.