Automated methods and systems for construction planning and scheduling: Critical review of three decades of research

Over the last 3 decades, a large body of research focused on automated construction planning and scheduling. Some of these efforts introduced methods to use design information to automatically develop the scope of work, establish work breakdown structures, and create optimal project sequences. Others introduced new techniques to formalize the sequencing relationships among schedule activities and project components. Despite these advancements, most construction projects - if not all - still are engaged fully in manual workflows of planning and scheduling. By offering a critical review of the literature, this manuscript examines the key issues that, to date, have hindered scaling and wide adoption of automated planning methods and systems. A close examination of how knowledge is formalized; scope quantification and project definition methods; and planning, scheduling, and schedule optimization techniques identified the following gaps in knowledge: (1) lack of flexibility in how construction knowledge is stored in existing construction method model templates for sequencing algorithms; (2) the dependency of current automated scheduling methods on manually formed and maintained work templates; (3) lack of learning methods to automate learning of construction knowledge from existing records without extensive human input; (4) limited validation on applicability of existing automated planning systems on real-life construction projects; and (5) the decoupled nature of research on automated planning versus schedule optimization. Building on the recent advancements in deep learning and natural language processing and the rise in adoption of lean construction theories, a discussion is offered on the path for research toward automatic generation of dynamic work templates and their inclusion in integrated planning, scheduling, and optimization systems.