Scholarlink

In this paper, we have presented a cost prediction (CP) model is used to predict the cost based on the historical information of the construction project. To accomplish this goal, in this research, inclusive artificial neural network (ANN) is used and performance of it is enhanced using the metaheuristic group learning algorithm (GLA) by finding the best weight values. The GLA algorithm is chosen over other algorithms due to better access, exploration and exploitation rate. Further, in this research, we have designed a multi-objective function by considering the MSE and RMSE parameters. The proposed model is simulated on the standard dataset of GAZA construction project and evaluated using the parameters, namely, MSE, RMSE, and MAPE. The result shows that the proposed model is efficiently predict the cost with respect to the actual cost. Finally, the comparative analysis shows that the proposed model achieves MAPE value of 25.972 and RMSE 116490 which is lesser than ANN model.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cost Prediction Model Based on Artificial Neural Network and Group Learning Algorithms for Construction Projects

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ravi Gaba,
  • Neeru Singla,
  • Rinku Walia,
  • Ajay Kumar,
  • Sandeep Singla

摘要

In this paper, we have presented a cost prediction (CP) model is used to predict the cost based on the historical information of the construction project. To accomplish this goal, in this research, inclusive artificial neural network (ANN) is used and performance of it is enhanced using the metaheuristic group learning algorithm (GLA) by finding the best weight values. The GLA algorithm is chosen over other algorithms due to better access, exploration and exploitation rate. Further, in this research, we have designed a multi-objective function by considering the MSE and RMSE parameters. The proposed model is simulated on the standard dataset of GAZA construction project and evaluated using the parameters, namely, MSE, RMSE, and MAPE. The result shows that the proposed model is efficiently predict the cost with respect to the actual cost. Finally, the comparative analysis shows that the proposed model achieves MAPE value of 25.972 and RMSE 116490 which is lesser than ANN model.