Creating quality construction schedules to mitigate delays often relies on experience. The lack of dissemination of historic decision reasoning makes it harder. This study proposes a graph-based method to find the time- and risk-efficient construction method patterns from historic projects to help schedulers improve productivity and accuracy. The method leverages schedule data obtained from a Tier-1 contractor and validates for excavation activities. The results indicate that the most time-efficient excavation activities can be done in 0.6% of total project time. The proposed method can help industry professionals standardise scheduling guidelines and automate the generation of construction schedules for critical subtasks.