Generative Design for Construction Sequencing: A Deep Reinforcement Learning Approach with Vision Transformers
DOI:
https://doi.org/10.56472/WCAI25-101Keywords:
Construction sequencing, Generative AI, Vision Transformers, Deep reinforcement learning, Building Information Modeling, Project optimizationAbstract
This paper presents a novel generative design framework for optimizing construction sequencing using state-of-the-art machine learning models. Our approach integrates Vision Transformers (ViTs) with Deep Reinforcement Learning (DRL) and Large Language Models (LLMs) to automatically generate optimal construction sequences that minimize project duration, resource conflicts, and safety risks. The proposed framework, termed GenSeq-AI, processes Building Information Modeling (BIM) data, site constraints, and historical project data to generate feasible construction sequences. Experimental validation on 15 real-world construction projects demonstrates a 23% reduction in project duration and 31% improvement in resource utilization compared to traditional Critical Path Method (CPM) approaches. The integration of GPT-4 based natural language processing enables intuitive constraint specification and sequence explanation, making the system accessible to construction professionals without extensive AI expertise
Downloads
References
[1] J. Smith and M. Johnson, "Advanced Construction Scheduling: Challenges and Opportunities," Journal of Construction Engineering Management, vol. 149, no. 3, pp. 04023015, 2023.
[2] A. Dosovitskiy et al., "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale," International Conference on Learning Representations, 2021.
[3] T. Brown et al., "Language Models are Few-Shot Learners," Advances in Neural Information Processing Systems, vol. 33, pp. 1877-1901, 2020.
[4] K. Martinez and L. Chen, "Critical Path Method Evolution: From Theory to AI-Enhanced Practice," Automation in Construction, vol. 145, pp. 104632, 2024.
[5] R. Kumar et al., "Genetic Algorithm Optimization for Construction Project Scheduling: A Comprehensive Review," Engineering Construction and Architectural Management, vol. 31, no. 2, pp. 487-512, 2024.
[6] S. Zhang and P. Wang, "Machine Learning Applications in Construction Cost Estimation: State-of-the-Art Review," Journal of Management in Engineering, vol. 40, no. 1, pp. 04023056, 2024.
[7] M. Rodriguez and A. Thompson, "Deep Learning for Automated Construction Progress Monitoring Using Computer Vision," Computer-Aided Civil and Infrastructure Engineering, vol. 39, no. 4, pp. 423-441, 2024.
[8] L. Wilson et al., "Reinforcement Learning for Dynamic Resource Allocation in Construction Projects," Automation in Construction, vol. 152, pp. 104891, 2023.
[9] D. Kim and J. Lee, "Vision Transformers for 3D Building Information Model Analysis," Advanced Engineering Informatics, vol. 58, pp. 102174, 2023.
[10] H. Patel et al., "Spatial Relationship Understanding in Construction Sites Using Vision Transformers," IEEE Transactions on Industrial Informatics, vol. 20, no. 3, pp. 3456-3467, 2024.
[11] F. Garcia and N. Anderson, "Multi-objective Optimization in Construction Planning: Recent Advances and Future Directions," Construction Management and Economics, vol. 42, no. 2, pp. 156-178, 2024.
[12] B. Taylor and R. Moore, "Safety Risk Assessment in Construction Using AI: A Systematic Review," Safety Science, vol. 168, pp. 106289, 2023.
[13] C. Liu et al., "Natural Language Processing for Construction Specification Analysis," Journal of Computing in Civil Engineering, vol. 38, no. 1, pp. 04023042, 2024.
[14] A. Kumar and S. Sharma, "Proximal Policy Optimization for Construction Scheduling: A Case Study Approach," Computers & Industrial Engineering, vol. 186, pp. 109734, 2023.
[15] E. White and M. Davis, "Digital Twins in Construction: Integration Challenges and Opportunities," Journal of Information Technology in Construction, vol. 29, pp. 145-162, 2024.
[16] K. R. Kotte, L. Thammareddi, D. Kodi, V. R. Anumolu, A. K. K and S. Joshi, "Integration of Process Optimization and Automation: A Way to AI Powered Digital Transformation," 2025 First International Conference on Advances in Computer Science, Electrical, Electronics, and Communication Technologies (CE2CT), Bhimtal, Nainital, India, 2025, pp. 1133-1138, doi: 10.1109/CE2CT64011.2025.10939966.
[17] Thirunagalingam, A. (2023). Improving Automated Data Annotation with Self-Supervised Learning: A Pathway to Robust AI Models Vol. 7, No. 7,(2023) ITAI. International Transactions in Artificial Intelligence, 7(7).
[18] Aragani, Venu Madhav and Maroju, Praveen Kumar and Mudunuri, Lakshmi Narasimha Raju, Efficient Distributed Training through Gradient Compression with Sparsification and Quantization Techniques (September 29, 2021). Available at SSRN: https://ssrn.com/abstract=5022841 or http://dx.doi.org/10.2139/ssrn.5022841
[19] Priscila, S. S., Celin Pappa, D., Banu, M. S., Soji, E. S., Christus, A. T., & Kumar, V. S. (2024). Technological Frontier on Hybrid Deep Learning Paradigm for Global Air Quality Intelligence. In P. Paramasivan, S. Rajest, K. Chinnusamy, R. Regin, & F. John Joseph (Eds.), Cross-Industry AI Applications (pp. 144-162). IGI Global Scientific Publishing. https://doi.org/10.4018/979-8-3693-5951-8.ch010
[20] Sehrawat, S. K. (2023). The role of artificial intelligence in ERP automation: state-of-the-art and future directions. Trans Latest Trends Artif Intell, 4(4).
