Publications / 2024 Proceedings of the 41st ISARC, Lille, France
Traditional construction management practices often suffer from inefficiencies and waste, leading to increased costs and project delays. Lean Construction (LC), by emphasizing waste reduction and customer value prioritization, has emerged as an effective construction management system. In recent years, efficiency enhancement approaches such as Lean are being further augmented by automation and AI tools in the Architecture, Engineering, and Construction (AEC) industry.
This work develops an ML-based image processing algorithm that can be used within the Lean Construction framework. This algorithm classifies construction images based on the 5S Lean tool's sort and set in order principles, utilizing image analysis techniques. The algorithm that leverages convolutional neural networks and TensorFlow/Keras libraries, shows promise in effective image classification under the 5S framework. This study also proves that YOLO models can integrate lean concepts with AI Tools to foster improvements in productivity, safety, and sustainability in the construction industry.