Abstract:
One of the biggest challenges facing the construction industry is project control and planning. This is because uncertainties, unrealistic planning, and other reasons. With the rise of AI language models and technologies such as ChatGPT, which has impressed people with its articulate answers to a wide range of questions, a unique opportunity emerges for traditional project control methods to be improved and transformed.
We developed an AI language model that learned from over 5,000 project schedules to identify best practices. Our analysis revealed 236 frequent sequences and 90 valid frequent sequences related to onsite work. Productivity of these sequences varies, with earthworks being the most time-consuming, especially in road and heavy construction projects.
The results of this project are expected to reduce the risk of project delays caused by unrealistic and poor planning. This could lead to a reduction in construction costs, rework, and associated material waste. Less experienced project managers can rely on the insights provided by machine learning, rather than relying solely on the experiences of seasoned project managers who may have learned from past failures.