Work-related overexertion injuries in cleaning occupations: An exploration of the factors to predict the days of absence by means of machine learning methodologies

Aroa González Fuentes; Nélida M Busto Serrano; Fernando Sánchez Lasheras; Gregorio Fidalgo Valverde; Ana Suárez Sánchez

doi:10.1016/j.apergo.2022.103847

Work-related overexertion injuries in cleaning occupations: An exploration of the factors to predict the days of absence by means of machine learning methodologies

Appl Ergon. 2022 Nov:105:103847. doi: 10.1016/j.apergo.2022.103847. Epub 2022 Jul 30.

Authors

Aroa González Fuentes¹, Nélida M Busto Serrano², Fernando Sánchez Lasheras³, Gregorio Fidalgo Valverde⁴, Ana Suárez Sánchez⁵

Affiliations

¹ School of Mining, Energy and Materials Engineering of Oviedo, University of Oviedo, Spain. Electronic address: UO212081@uniovi.es.
² Labor and Social Security Inspectorate. Ministry of Labor and Social Economy, Spain. Electronic address: nelida.busto@mites.gob.es.
³ Department of Mathematics, University of Oviedo, 33007, Oviedo, Spain; Instituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), University of Oviedo, 33004, Oviedo, Spain. Electronic address: sanchezfernando@uniovi.es.
⁴ Instituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), University of Oviedo, 33004, Oviedo, Spain; Department of Business Management, University of Oviedo, 33004, Oviedo, Spain. Electronic address: gfidalgo@uniovi.es.
⁵ Instituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), University of Oviedo, 33004, Oviedo, Spain; Department of Business Management, University of Oviedo, 33004, Oviedo, Spain. Electronic address: suarezana@uniovi.es.

PMID: 35917696
DOI: 10.1016/j.apergo.2022.103847

Abstract

The special characteristics of the cleaning industry have an important impact on the health and safety of its workforce. Making use of data from more than 79,000 occupational accidents, the aim of the present research is to use machine learning techniques to develop a model to predict incapacity for work (expressed in days of absence) due to work-related overexertion injuries among service sector cleaners in Spain. The severity of accidents caused by overexertion depends on several factors that can be classified into the following categories: injury typology, individual factors, employment conditions, accident circumstances and health and safety management and standards in the company.

Keywords: Absenteeism; Cleaning sector; Machine learning; Musculoskeletal disorders (MSD); Work-related overexertion injuries.