ITcon Vol. 30, pg. 650-678, http://www.itcon.org/2025/27

Automatic detection of the health status of workplaces by processing building end-users’ maintenance requests

DOI:10.36680/j.itcon.2025.027
submitted:October 2024
revised:April 2025
published:May 2025
editor(s):Turk Ž
authors:Marco D’Orazio, Prof.
DICEA, Università Politecnica delle Marche, via brecce bianche, Ancona, Italy
ORCID: https://orcid.org/0000-0003-3779-4361
m.dorazio@staff.univpm.it

Gabriele Bernardini, Dr.
DICEA, Università Politecnica delle Marche, via brecce bianche, Ancona, Italy
ORCID: https://orcid.org/0000-0002-7381-4537
g.bernardini@staff.univpm.it

Elisa Di Giuseppe, Prof.
DICEA, Università Politecnica delle Marche, via brecce bianche, Ancona, Italy
ORCID: https://orcid.org/0000-0003-2073-1030
e.digiuseppe@staff.univpm.it
summary:This paper addresses the challenge of assessing workplace health through building maintenance requests’ data, particularly focusing on the impact of maintenance conditions on workers' satisfaction, well-being and possible stress levels. A data-driven methodology based on CMMS (Computer Management Maintenance Systems) is proposed, utilizing indexes to measure both the quantity and perceived quality of maintenance interventions. Sentiment and emotion analysis, along with lexical diversity indices, are applied to capture the perceptions of end-users and technical staff. The methodology successfully identifies maintenance issues in buildings and highlights differences in perception between workers' typologies. The results provide valuable insights for facility managers and organizations, enabling better-informed decisions on maintenance priorities based on both objective data and workers' feedback. This approach paves the way for future research integrating qualitative and quantitative data in facility management, with the potential to enhance decision-making and improve workplace health.
keywords:Building maintenance, sentiment and emotion analysis, natural language processing, user perception, workplace health status, data-driven approach
full text: (PDF file, 2.173 MB)
supplementary material:
citation:D’Orazio M, Bernardini G, Di Giuseppe E (2025). Automatic detection of the health status of workplaces by processing building end-users’ maintenance requests, ITcon Vol. 30, pg. 650-678, https://doi.org/10.36680/j.itcon.2025.027
statistics: