Hybrid approach for cost estimation of sustainable building projects using artificial neural networks

Abstract: Inaccurate estimation in sustainable construction projects is a significant challenge for appraisers, particularly when data and knowledge about the projects are lacking. As a result, there is a need to use cutting-edge technology to solve the issue of estimation inaccuracy. Iraq’s productivity estimates are now made using outdated, ineffective methodologies and procedures. In addition, it is essential to implement cutting-edge, quick, precise, and adaptable technology for productivity estimation. This study’s major goal is to calculate the overall costs of sustainable buildings using the cutting-edge technique known as artificial neural networks (ANNs). For Iraq’s construction industry to handle projects successfully, ANNs must be used as a new technology, a methodology developed to estimate the overall costs of sustainable construction projects. In this study, the process of cost estimation was modeled using ANNs. Investigations of a number of examples involving the creation of ANNs have also been made, including network design and internal elements and how much they impact the effectiveness of models built using ANNs. Equations were developed to determine structural productivity. These networks were shown to have extremely strong predictive power for both accounting coefficients (R) (93.33%) and the overall costs of sustainable construction projects, with a prediction accuracy of 87.00 and 93.33%, respectively.

Angaben zum Objekt

• Standort: Deutsche Nationalbibliothek Frankfurt am Main
• Umfang: Online-Ressource
• Sprache: Englisch

Verweise und Beziehungen

• Erschienen in: Hybrid approach for cost estimation of sustainable building projects using artificial neural networks ; volume:14 ; number:1 ; year:2024 ; extent:8 ; Open engineering ; 14, Heft 1 (2024) (gesamt 8)

Beteiligte, Orts- und Zeitangaben

• Urheber: Al-Somaydaii, Jumaa A. ; Albadri, Aminah T. ; Al-Zwainy, Faiq M. S.

Weitere Informationen

• DOI: 10.1515/eng-2022-0485
• URN: urn:nbn:de:101:1-2024032114462790675290
• Rechteinformation: Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
• Letzte Aktualisierung: 14.08.2025, 10:58 MESZ