An Artificial Neural Networks with Particle Swarm Optimization using ReliefF for Construction Cost Evaluation

N ow a days, construction cost evaluation is a dynamic process in project management which enables accurate estimation by analysis and controlling of costs to guarantee projects are completed efficiently, within budget, and to desired quality standards. Blasting is a crucial aspect of the construction and mining industries. However, blast designers face a persistent challenge of minimizing the adverse effects of explosive usage on surrounding areas, particularly flyrock and ground vibration which has unintended consequences on the environment and nearby structure. To overcome these issues Artificial Neural N etworks-Particle Swarm Optimization (ANN-PSO) is proposed by collecting the input from the Construction Spending dataset. The ReliefF algorithm selects relevant features from the dataset, which are then used as input for classification. The proposed method hybridizes ANN-PSO classification improve accuracy and efficiency of the construction cost evaluation. The ensemble method leverages the strengths of both techniques, providing a robust approach for classification and prediction in construction spending. The proposed ANN-PSO produce more results than traditional methods of Particle Swarm Optimization Deep Neural Networks (PSO-DNN) for modeling extreme values such as R2 and Root Mean Squared Error (RMSE) with the values of (0.003) and (0.067) respectively.