Predictive Modeling: BIM Command Recommendation Based on Large-scale Usage Logs

2025-02-23

arXiv.org

AI Summary

BIM adoption in the AEC industry is hindered by high operational complexity and low modeling efficiency. To address this, we propose the first real-time command recommendation framework tailored for BIM interaction logs, trained on 32 billion anonymized user operation records. Our approach employs a sequence recommendation model integrating an LLM-inspired Transformer architecture. Key innovations include a BIM-specific log filtering and enhancement strategy, a customized multi-source feature fusion module, a dedicated loss function, and a self-supervised pretraining–task-adaptive fine-tuning paradigm—enabling robust cross-national, cross-disciplinary, and cross-project generalization. Evaluated on real-world Vectorworks interaction logs, our framework achieves 84% Recall@10, demonstrating substantial improvements in modeling efficiency and strong generalization capability across diverse professional and project contexts.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The adoption of Building Information Modeling (BIM) and model-based design within the Architecture, Engineering, and Construction (AEC) industry has been hindered by the perception that using BIM authoring tools demands more effort than conventional 2D drafting. To enhance design efficiency, this paper proposes a BIM command recommendation framework that predicts the optimal next actions in real-time based on users' historical interactions. We propose a comprehensive filtering and enhancement method for large-scale raw BIM log data and introduce a novel command recommendation model. Our model builds upon the state-of-the-art Transformer backbones originally developed for large language models (LLMs), incorporating a custom feature fusion module, dedicated loss function, and targeted learning strategy. In a case study, the proposed method is applied to over 32 billion rows of real-world log data collected globally from the BIM authoring software Vectorworks. Experimental results demonstrate that our method can learn universal and generalizable modeling patterns from anonymous user interaction sequences across different countries, disciplines, and projects. When generating recommendations for the next command, our approach achieves a Recall@10 of approximately 84%.

Problem

• Enhance BIM design efficiency by predicting next actions
• Process large-scale BIM log data for command recommendations
• Improve command prediction accuracy using Transformer-based models

Innovation

• BIM command recommendation using Transformer backbones
• Custom feature fusion and targeted learning strategy
• Processes 32 billion rows of real-world BIM logs