A mixed integer programming method for multi-project resource leveling

Altun M., Sönmez R., Akçamete Güngör A.

Journal of Construction Engineering, Management Innovation, vol.3, no.2, pp.131-140, 2020 (Peer-Reviewed Journal)


In the implementation of construction projects, efficient resource planning plays a prominent role in developing cost-efficient solutions. Therefore, the decision makers should level the project resources with respect to the planned project schedule to reduce the project costs. In general, they usually focus on each project separately to optimize the project’s resource usage according to the intended resource objective function of the resource leveling problem. However, in real life, multiple projects with shared resources may be executed simultaneously. Hence, separate evaluations of resource leveling problem for each project may result in sub-optimal solutions due to negligence of the effects of the shared resources for the projects. Therefore, the projects using shared resources should be leveled together to reach the global optimum solution. In this study, an optimization model is developed using Mixed Integer Programming (MIP) to minimize peak requirements of different resource types in multiple construction projects. The performance of the proposed model is tested with four case study projects with different project settings. The solutions that are achieved with the proposed method are compared with the optimum solutions of the traditional leveling models which optimize the resource usage for each project separately. The results show that the proposed method considerably improves portfolio performance compared to the results of the traditional method based on separate leveling of projects.