Machine learning and multi-objective optimization methodology for planning construction phases of airport expansion projects

This study presents the development of an innovative methodology to optimize phasing plans of airport expansion projects to minimize construction -related disruptions in airport operations and project construction cost. This methodology integrates a machine learning model to predict the impact of generated phasing plans on flights ground movement time during construction, and a multi -objective genetic algorithms optimization model to identify optimal construction phasing plans for airport expansion projects. An airport expansion case study is analyzed to demonstrate the capabilities of the developed models. The results of this analysis confirm the original contributions of the novel methodology in predicting the impact of alternative construction phasing plans on flights ground movement time without the need for repetitive and time-consuming simulations, quantifying and optimizing the impact of alternative construction phasing plans on airport operations disruption cost and project construction cost, and generating detailed optimum phasing plans for airport expansion projects. These novel capabilities are expected to enable planners improving cost-effectiveness of airports during expansion projects while minimizing construction cost.