In construction, planning often involves complex and interconnected processes that requires managing numerous factors and multiple variables to get the planning right. Factors like location, workforce, product, finance, equipment, design, or external; play vital role in scheduling. Additionally, project variables like budget, logical sequencing of the activities, allocating and managing resources, efficiency of the project team, priorities of the stakeholders, risks and external factors makes the planning even more complex. Coming from the industry 4.0 framework, Artificial Intelligence (AI) leverages the purposefully developed intelligence machines or tools to carry out the complex calculation, multi-dimensional simulations, and pattern recognitions to solve similar problems. Therefore, the dynamic and compound complexity of planning is now proposed to be simplified using Machine Learning (ML) a subsidiary branch of AI. Importantly, the larger and more relevant dataset one has, the more accurate and logical patterns one can identify. Though there are few solutions available that utilizes ML for planning, the commonly used prediction validation labels in such prevailing prediction systems are more focused on Earned Value Management (EVM) and productivity which are solely focused to budgeting, sequencing, and resource management. Contrarily, the reliable and accurate system would fundamentally require inclusion of more reliable performance indices and different approach for data sources and collection. Accordingly, this study is aimed at developing such lean driven prediction engine.
