Graph Theory Implementation Towards Artificial Intelligence Generated Construction Schedule

9 pages•Published: May 26, 2024

Abstract

Commercial construction scheduling software (such as Primavera Project Planner and Microsoft Project) has been used by construction professionals for close to 40 years. Although those scheduling tools have been significantly modified over the years, the fundamental basis of the interaction between the construction professional and the scheduling software has remained unchanged where the construction professionals enter information and the software makes calculations and displays the schedule information in tables and graphs (such as CPM Diagram and Gantt Chart). However, in the same period, significant computational advances have been achieved leading to the recent explosion of Artificial Intelligence (AI) capabilities. The problem that serves as a point of departure for this research is that the current construction scheduling software workflow is unable to fully benefit from AI capabilities. Thus, the objective and original contribution of this paper is to describe the implementation of graph theory in Python to facilitate the realization of AI-generated construction schedules. The results presented in this paper are very encouraging as a piece in the puzzle of elements needed to create a new generation of construction scheduling software could benefit from AI capabilities and assist construction professionals in the creation of the construction schedule beyond simply making calculations and displaying information.

Keyphrases: Node Anomaly, PERT Diagram, PERT Motif, Relationship Prediction, Weighted PERT

In: Tom Leathem, Wesley Collins and Anthony J. Perrenoud (editors). Proceedings of 60th Annual Associated Schools of Construction International Conference, vol 5, pages 549--557

Links:	https://easychair.org/publications/paper/D32X
	https://doi.org/10.29007/4j4v