Have a personal or library account? Click to login
Building safe organisations: using machine learning to decode safety habits of blue-collar workers in the construction industry Cover

Building safe organisations: using machine learning to decode safety habits of blue-collar workers in the construction industry

Open Access
|Apr 2026

Abstract

This study aims to provide a framework for categorising safety behaviours of construction workers, recognising the importance of employee safety in the competitive business environment. Employee safety is crucial to overall efficiency, productivity, and well-being, and the study seeks to contribute to understanding and managing workplace safety in the construction industry.

This study utilises machine learning (ML) algorithms, like logistic regression, support vector machine, and decision trees, to develop a categorisation framework for the safety behaviours of construction workers. The framework is validated using frequent safety behaviours observed in a random sample of construction professionals.

The study finds that workplace safety behaviours (WSB) are primarily influenced by supervisor support, reckless habits, and safety motivation. Limiting workplace accidents, enforcing safety laws, properly documenting safety processes, and organising sessions to educate staff are identified as critical sub-factors. Advancements in technology have resulted in significant improvements across construction organisations in allied domains. Additional considerations include education, preempting the possibility of accidents in different workplace situations, and enforcing strong disciplinary measures.

The framework proposed can serve as a valuable tool for organisations to tailor safety interventions. By recognising the diverse influences on safety behaviours, companies can implement targeted measures to address specific root causes of unsafe practices. The practical implications of these findings for safety management in the construction industry are noteworthy.

DOI: https://doi.org/10.2478/emj-2026-0004 | Journal eISSN: 2543-912X | Journal ISSN: 2543-6597
Language: English
Page range: 42 - 59
Submitted on: Jul 1, 2025
|
Accepted on: Dec 15, 2025
|
Published on: Apr 2, 2026
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2026 Arpit Singh, Ashish Dwivedi, Malini Mittal Bishnoi, Swamynathan Ramakrishnan, Dragan Pamucar, Anchal Patil, published by Bialystok University of Technology
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.