Detection and In-Depth Analysis of Causes of Delay in Construction Projects: Synergy between Machine Learning and Expert Knowledge

Marija Z. Ivanović (), Đorđe Nedeljković, Zoran Stojadinović, Dejan Marinković, Nenad Ivanišević and Nevena Simić
Additional contact information
Marija Z. Ivanović: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia
Đorđe Nedeljković: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia
Zoran Stojadinović: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia
Dejan Marinković: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia
Nenad Ivanišević: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia
Nevena Simić: Department of Construction Project Management, Faculty of Civil Engineering, University of Belgrade, 11000 Belgrade, Serbia

Sustainability, 2022, vol. 14, issue 22, 1-23

Abstract: Due to numerous reasons, construction projects often fail to achieve the planned duration. Detecting causes of delays (CoD) is the first step in eliminating or mitigating potential delays in future projects. The goal of research is unbiased CoD detection at a single project level, with the ultimate goal to discover the root causes of delay. The existing approach is based on expert knowledge which is used to create CoD lists for projects in general or groups of similar projects. When applied to a single project, it is burdened with bias, as shown on a case project returning low Spearman Rank correlation values. This research introduces a Delay Root causes Extraction and Analysis Model—DREAM. The proposed model combines expert knowledge, machine learning techniques, and Minutes of Meetings (MoM) as an unutilized extensive source of information. In the first phase, DREAM outputs a CoD list based on occurring frequency in MoM with satisfactory recall values, significantly reducing expert-induced subjectivism. In the second phase, enabled by MoM dates, DREAM adds another dimension to delay analysis—temporal CoD distribution. By analyzing corresponding informative charts, experts can understand the nature of delays and discover the root CoD, allowing intelligent decision making on future projects.

Keywords: causes of delay; machine learning; transformers; bias; Spearman rank correlation; construction projects (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 (1)

Downloads: (external link)
https://www.mdpi.com/2071-1050/14/22/14927/pdf (application/pdf)
https://www.mdpi.com/2071-1050/14/22/14927/ (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:22:p:14927-:d:969890

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 ().