EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Research on Engineering Cost Estimation Based on Enhanced Neural Network Optimization Method

Jiaqi Xu

Engineering Reports · 2026-07 · 8 (7)

DOI: 10.1002/eng2.70802 ↗

At a glance

Journal article in Engineering Reports, published online on 2026-07-08. Written by Jiaqi Xu. 29 references. The full text is freely available (open access, CC BY 4.0).

AbstractBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Considering the fundamental challenge of accurately and reliably predicting the final cost of engineering projects, due to the multiple factors and the complexity of the relationships between them, the present study presents an innovative framework. This framework, with the aim of improving the accuracy of the predictions, is based on hybrid feature selection and advanced neural network optimization using genetic algorithms guided by reinforcement learning. The proposed method consists of three main phases: In the first phase, preprocessing of raw records is performed to obtain a set of suitable features for processing. In the second step, a hybrid feature selection algorithm based on information gain and Boruta is used. These techniques simultaneously rank the features, and after normalizing the weights and determining the final ranking, the Recursive Feature Elimination with Cross‐Validation (RFE‐CV) is used to select the most relevant features. In the third phase, to perform the prediction operation, a multilayer perceptron (MLP) neural network is optimized using a new optimization algorithm called Q‐Learning‐assisted Genetic Algorithm (QLGA). By adjusting the structure and weight vector, an efficient model is created according to the training data, which is able to improve the prediction performance compared to conventional training algorithms. In this optimization algorithm, the Q‐Learning strategy is used to guide the population towards the global optimum. After determining the optimal MLP model, it is used to predict new samples. The root mean squared error and mean absolute percentage error values of 8.2 and 4.5, respectively, show that the proposed method is very effective at predicting the final costs of engineering projects compared to other methods that are already out there.

Cite

APA

Xu, J. (2026). Research on Engineering Cost Estimation Based on Enhanced Neural Network Optimization Method. Engineering Reports, 8(7). https://doi.org/10.1002/eng2.70802

Vancouver

Xu J. Research on Engineering Cost Estimation Based on Enhanced Neural Network Optimization Method. Engineering Reports. 2026;8(7). doi:10.1002/eng2.70802

BibTeX
@article{Xu2026,
  author = {Xu, Jiaqi},
  title = {Research on Engineering Cost Estimation Based on Enhanced Neural Network Optimization Method},
  journal = {Engineering Reports},
  year = {2026},
  volume = {8},
  number = {7},
  doi = {10.1002/eng2.70802}
}
Start for free →