Hybrid Genetic Algorithm and Constraint-Based Simulation Framework for Building Construction Project Planning and Control

Abstract

In the construction sector, the absence of a framework to incorporate and synchronize the resources required for activities means that projects risk incurring cost and time overruns. In this study, a framework was generated and tested to optimize resource allocation. The framework considers financial constraints through a genetic algorithm process and constraint-based simulation. The simulation method was developed based on the critical path method relation-defining system, with a dynamic prioritizing system automatically creating the simulation network. By introducing a standard cost database to the data management section of the framework, the total work required to insert data manually is minimized. For the majority of the activities, the user simply chooses an activity and determines its relationships with other activities. The rest of the initial required data are automatically pulled from the imported databases. The validity of the developed framework was demonstrated through a case study on constructing the structure of a residential building in Tehran, including four structural systems. The four systems were compared regarding time and cost of execution to illustrate the relevance of the framework. The framework decreases the user’s manual work while producing outputs of time, cost, and resources that are compatible with building information modeling (BIM) software. Overall, the result is an enhanced quality of project management plans. Moreover, the framework can be used as a planning tool at the preconstruction stage or as a control tool during the construction phase. The proposed approach provides a realistic project management plan, with the potential for substantial savings in finances and resources.