AUTOMATING CONSTRUCTION SUPPLY CHAIN USING ML: REVOLUTIONZNING PAKISTAN’S CONSTRUCTION SUPPLY CHAIN WITH AI AND ML

Abstract

Supply chain management in the construction industry is very critical phenomenon. However, challenges in the supply chain often led to delays, cost overruns, and quality issues in projects. This research focuses on addressing the significant challenges faced by the construction industry in Pakistan's Supply Chain Management, particularly in the realm of material estimation, procurement inefficiencies, and resource allocation. The study adopted a quantitative iterative experimental approach by utilizing Machine Learning AI, namely YOLO V8 and python coding to automate the process of Cost Estimation in construction supply chain. The study aims to serve as a proof for concept that “Detection using ML algorithms” can be used to automate Cost Estimation. The YOLOv8 model was trained on 5 blueprint components (walls, floor, columns, windows, and doors). Python codes were used to extract bounding box coordinates which were further used for cost estimation. The research findings show data that yielded high accuracy percentages, establishing that with better ML training sets, 100% accuracy of cost estimations is achievable.