A Metaheuristic Procedure Combined with 4D Simulation as an Alternative for the Scheduling Process of Housing Complexes

Abstract

The present work focuses on the application of a metaheuristic procedure, represented by a Genetic Algorithm, in the search of better alternatives for the organization of activities in the construction of housing complexes. Therefore, the algorithm is implemented in order to solve the Resource-Constrained Project Scheduling Problem – RCPSP. The problem modeling is adapted to the type of project in question, where, at first, the activity network of a single housing unit is defined through an activity-on-the-node graph and afterwards the Line of Balance method is used in order to determine the start and finish dates of all the other repetitive activities of the project. In order to test the efficiency of the algorithm, numerous experiments are performed with project instances from the PSPLIB, a specific benchmark database for the RCPSP, available online. Finally, in order to better visualize the schedules obtained in the optimization phase, a 4D simulation of a schedule for the execution of a housing complex, generated with the metaheuristic procedure, is performed with the software Autodesk Navisworks. The simulation phase allows the visualization of the projects constructability, makes it possible to confirm if the algorithm-generated schedule follows a logical sequence, as well as it supports the decision-making about the best execution alternatives.