Performance analysis of simulation-based optimization of construction projects using High Performance Computing

The complexity and uncertain nature of bridge construction projects require simulation for analyzing and planning these projects. On the other hand, optimization can be used to address the inverse relationship between the cost and time of a project and to find a proper trade-off between these two key elements. In addition, the large number of resources required in large-scale bridge construction projects results in a very large search space. Therefore, there is a need for using parallel computing to reduce the computational time of the simulation-based optimization problem. Another problem in this area is that most of the construction simulation tools need an integration platform to be combined with optimization techniques. To alleviate these limitations, an integrated simulation-based optimization framework is developed within one High Performance Computing (HPC) platform, and its performance is analyzed by carrying out a case study. A master-slave (or global) parallel Genetic Algorithm (GA) is used to decrease the computation time and to efficiently use the full capacity of the computer. In addition, sensitivity analysis is applied to identify the promising configuration for GA and the best number of cores used in parallel and to analyze the impact of GA parameters on the overall performance of the simulation-based optimization model. Using NSGA-ΙΙ as the optimization algorithm resulted in better near-optimal solutions compared to those of fast-messy GA. Moreover, performing the proposed framework on multiple nodes using the cluster system led to 31% saving in the computation time on average. Furthermore, the GA was tuned using sensitivity analyses, which resulted in the selection of the best parameters of the GA.