Original scientific paper
https://doi.org/10.2478/otmcj-2025-0004
An artificial neural network model to relate organisation characteristics and delivery methods of construction projects
Moein Pashaian
; Department of Civil Engineering, Sari Branch, Islamic Azad University, Sari, Iran
Babak Fazli Malidareh
; Department of Civil Engineering, Babol Branch, Islamic Azad University, Babol, Iran
*
Seyedeh Mona Tabandeh
; Department of Civil Engineering, Babol Branch, Islamic Azad University, Babol, Iran
* Corresponding author.
Abstract
This paper presents an artificial neural network (ANN) model designed to predict the optimal delivery methods for construction projects based on organisational characteristics. Effective organisational characteristics were identified through a combination of the Delphi method and data collected via questionnaire surveys. The study sample consisted of 354 construction experts selected using a random sampling method. The validity and reliability of the research were confirmed through the formcontent validity and the Cronbach’s alpha test, respectively. The ANN model, implemented using RapidMiner software, demonstrated a prediction accuracy of 76.42%. The results revealed that financial, managerial, contextual, optimisation, and manpower variables significantly impact the prediction of the delivery method. Compared to other data mining models, such as the decision tree, random forest, and support vector machine (SVM), the ANN model showed a superior accuracy. This research highlights the contribution of organisational characteristics in forecasting the delivery methods of construction projects and offers a novel approach to improving project delivery decisions. While the findings are based on data from the Mazandaran province in Iran, the methodology and insights can be adapted and applied to other regions with similar organisational characteristics, suggesting a potential for generalisation.
Keywords
artificial neural network, organisational characteristics, delivery of construction projects
Hrčak ID:
342004
URI
Publication date:
1.1.2025.
Visits: 1 *