Comparative analysis between different risk score calculation approaches

Abstract

Purpose

Total risk score (TRS) is considered one of the main indicators for risk evaluation. Several studies attempted to employ different types of risk indices for the evaluation of cost overrun causes. Hence, this study aims at performing a comparative analysis to evaluate the efficiency of three different approaches for TRS calculation.

Design/methodology/approach

Thirty-eight unique causes of cost overrun in urban-related construction projects were identified and a survey was conducted among construction professionals in Iran. The TRS for each cost overrun cause is calculated using single-attribute (SA), double-attribute (DA), and multiple-attribute (MA) approaches, and eventually, causes were ranked. Furthermore, principal component analysis (PCA), logistic regression analysis (LRA), and K-means clustering are utilized to compare the differences in the generated TRS using different approaches.

Findings

The results revealed that the TRS generated through the MA approach demonstrated the highest efficiency in terms of generating correlation between causes and their identified latent constructs, prediction capability, and classification of the influential causes in the same group.

Originality/value

The originality of this study primarily stems from the adoption of statistical approaches in the evaluation of the recently introduced TRS calculation approach in comparison to traditional ones. Additionally, this study proposed a modified application of the relative importance index (RII) for risk prioritization. The results from this study are expected to fulfill the gap in previous literature toward exploring the most efficient TRS calculation approach for those researchers and practitioners who seek to utilize them as a measure to identify the influential cost overrun causes.

Citation

Youssefi, I. and Celik, T. (2024), "Comparative analysis between different risk score calculation approaches", Engineering, Construction and Architectural Management, Vol. 31 No. 10, pp. 4099-4124. https://doi.org/10.1108/ECAM-11-2022-1097