Abstract/Summary

The aim of this study is to develop an activity-based lessons-learned model that allows construction companies to capture, store, classify, and reuse activity-related lessons learned from previous projects, thereby increasing the reliability of time estimates in scheduling. Scheduling is a knowledge-intensive process that requires the utilization of data and expert opinion elicitation from various levels of an organization in construction projects. This research consists of five successive steps: performing a needs analysis, proposing an activity-based lessons-learned process model, validating the proposed process model, developing a tool to apply the proposed model in a computer environment, and testing the applicability of the tool. To implement the proposed model in practice, a web-based tool, namely the Construction Industry Scheduling with Activity-Based Lessons Learned Tool (ConSALL Tool), was developed. Its functionality was evaluated using black-box testing. The tool was then applied in a real construction project. Results show that ConSALL has the potential to improve scheduling decisions in construction projects by incorporating data and experience from previous projects. Findings from this research can be used to develop similar models and AI tools to foster activity-based learning in other project-based industries as well as the construction industry. This paper presents an innovative approach to enhancing construction project scheduling by leveraging lessons learned from past projects. The development and application of the ConSALL Tool demonstrate a practical implementation of the proposed model, providing a framework that can be adapted to other industries to improve project planning and execution.