Forecasting Liquidated Damages via Machine Learning-Based Modified Regression Models for Highway Construction Projects

Odey Alshboul, Mohammad A. Alzubaidi, Rabia Emhamed Al Mamlook, Ghassan Almasabha, Ali Saeed Almuflih and Ali Shehadeh
Additional contact information
Odey Alshboul: Department of Civil Engineering, Faculty of Engineering, The Hashemite University, Zarqa 13133, Jordan
Mohammad A. Alzubaidi: Department of Computer Engineering, Hijjawi Faculty for Engineering Technology, Yarmouk University, Shafiq Irshidatst, Irbid 21163, Jordan
Rabia Emhamed Al Mamlook: Department of Industrial Engineering, Al Zawiya University, Al-Zawiya City, Libya
Ghassan Almasabha: Department of Civil Engineering, Faculty of Engineering, The Hashemite University, Zarqa 13133, Jordan
Ali Saeed Almuflih: Department of Industrial Engineering, King Khalid University, King Fahad St, Guraiger, Abha 62529, Saudi Arabia
Ali Shehadeh: Department of Civil Engineering, Hijjawi Faculty for Engineering Technology, Yarmouk University, Shafiq Irshidatst, Irbid 21163, Jordan

Sustainability, 2022, vol. 14, issue 10, 1-25

Abstract: Sustainable construction projects are essential for economic and societal thriving in modern communities. However, infrastructural construction is usually accompanied by delays in project delivery, which impact sustainability. Such delays adversely affect project time, cost, quality, safety objective functions, and associated Liquidated Damages (LDs). LDs are monetary charges to recompense the owner for additional expenses sustained if the project was not delivered on time due to delays caused by the contractor. This paper proposes modified regression modeling using machine learning (ML) techniques to develop solutions to the problem of predicting LDs for construction projects. The novel modeling methodology presented here is based on six years of data collection from many construction projects across the United States. It represents an innovative use of Multiple Linear Regression (MLR) models hybridized with machine learning (ML). The proposed methodology is evaluated using real datasets, where the developed model is designed to outperform the state-of-the-art LD forecast accuracy. Herein, seven modified regression-based models showed high accuracy in predicting the LDs. Nevertheless, those models’ forecasting ability was limited, so another second-order prediction model is proposed to provide better LD estimations. Independent variables were categorized based on their influence on the estimated LDs. The Total Bid Amount variable had the highest impact, while the Funding Indicator variable had a minimal impact. LD prediction was negatively correlated with all change-order-related variables and Total Adjustment Days, which suggests that those variables introduce extreme uncertainties due to their complex nature. The developed prediction models help decision-makers make better LDs predictions, which is essential for construction project sustainability.

Keywords: sustainability; liquidity damages; construction project; multiple linear regression (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/2071-1050/14/10/5835/pdf (application/pdf)
https://www.mdpi.com/2071-1050/14/10/5835/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:14:y:2022:i:10:p:5835-:d:813383

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().