This paper presents the results of an interactive virtual reality (VR) training program aimed at enhancing Health and Safety (H&S) management practices in quarrying operations. The course was designed based on industry best practices, as well as both voluntary and mandatory standards relevant to marble mining activities. It combines experiential learning with a performance monitoring system that tracks completion rates, time taken, and scores based on user decisions. The primary objective was to assess the impact of VR training across different user groups, categorized by age, prior safety experience, familiarity with equipment and processes, and VR proficiency. This study involved 40 participants and analyzed 15 variables, including occupation, age, H&S skills, process knowledge, equipment familiarity, VR skills, physical impact of VR, number of attempts before completion, percentage and time of completion, achieved scores, retention of knowledge, and user feedback before and after training. Performance measurement was carried out using two methods: a Microsoft Forms questionnaire with 16 questions, completed by participants one week after training, and Simula Solution, which automatically tracked and recorded performance metrics (time, percentage, errors, and scores) during each session. The survey successfully identified which demographic groups were most affected by VR training. The findings of this study could have important implications for improving H&S practices in the mining sector by empowering workers to engage in training and interact with process resources. This allows them to experience virtual accidents in a controlled, risk-free environment.

This document proposes a new evaluation model to be applied to a training course on health and safety at work based on virtual reality. The model refers to three macro-levels (design, delivery, and evaluation), which extend throughout the training life cycle. At macro level 1, design, the quality of the model intended for the virtual reality experience is evaluated, as well as its adaptation to the work environment and its compliance with applicable voluntary and mandatory standards; in macro level 2, delivery, the performance of the model, the individual reactions of users with headsets, their performance and psycho-physical state, the time, and the score achieved are evaluated; in macro level 3, evaluation, the long-term effects of subjective training and the social and economic impact that virtual reality training has had on the organisation are evaluated. The study investigates assessment models for virtual-reality-based occupational health and safety courses and identifies a model outlining general criteria that can be adapted to several types of courses and different work sectors. By examining the typical stages of the training life cycle and drawing on training evaluation models such as Kirkpatrick or Molenda and Information and Communication Technology metrics, the study identifies the key elements for assessing the effectiveness of virtual reality training in occupational health and safety.

Increasingly, information technology facilitates the storage and management of data useful for risk analysis and event prediction. Studies on data extraction related to occupational health and safety are increasingly available; however, due to its variability, the construction sector warrants special attention. This review is conducted under the research programs of the National Institute for Occupational Accident Insurance (Inail). Objectives: The research question focuses on identifying which data mining (DM) methods, among supervised, unsupervised, and others, are most appropriate for certain investigation objectives, types, and sources of data, as defined by the authors. Methods: Scopus and ProQuest were the main sources from which we extracted studies in the field of construction, published between 2014 and 2023. The eligibility criteria applied in the selection of studies were based on the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA). For exploratory purposes, we applied hierarchical clustering, while for in-depth analysis, we used principal component analysis (PCA) and meta-analysis. Results: The search strategy based on the PRISMA eligibility criteria provided us with 63 out of 2234 potential articles, 206 observations, 89 methodologies, 4 survey purposes, 3 data sources, 7 data types, and 3 resource types. Cluster analysis and PCA organized the information included in the paper dataset into two dimensions and labels: “supervised methods, institutional dataset, and predictive and classificatory purposes” (correlation 0.97–8.18 × 10⁻¹; p-value 7.67 × 10⁻⁵⁵–1.28 × 10⁻²²) and the second, Dim2 “not-supervised methods; project, simulation, literature, text data; monitoring, decision-making processes; machinery and environment” (corr. 0.84–0.47; p-value 5.79 × 10⁻²⁵–-3.59 × 10⁻⁶). We answered the research question regarding which method, among supervised, unsupervised, or other, is most suitable for application to data in the construction industry. Conclusions: The meta-analysis provided an overall estimate of the better effectiveness of supervised methods (Odds Ratio = 0.71, Confidence Interval 0.53–0.96) compared to not-supervised methods.

Increasingly, information technology facilitates the storage and management of data useful for risk analysis and event prediction. Studies on data extraction related to occupational health and safety are increasingly available; however, due to its variability, the construction sector warrants special attention. This review is conducted under the research programmes of the National Institute for Occupational Accident Insurance (Inail). Objectives: The research question focuses on identifying which data mining (DM) methods, among supervised, unsupervised, and others, are most appropriate to be applied to certain investigation objectives, types, and sources of data, as defined by the authors. Methods: Scopus and ProQuest were the main sources from which we extracted studies in the field of construction, published between 2014 and 2023. The eligibility criteria applied in the selection of studies, were based on the Preferred Reporting Items for Systematic Review and meta-analysis (PRISMA). For exploratory purposes, we applied hierarchical clustering, while for in-depth analysis, we use principal component analysis (PCA) and meta-analysis. Results: The search strategy based on the PRISMA eligibility criteria, provided us with 61 out of 2,234 potential articles, 202 observation, 91 methodologies, 4 survey purposes, 3 data sources, 7 data types, and 3 resource type. Cluster analysis and PCA organized the information included in the paper dataset into two dimensions and labels: "supervised methods, institutional dataset, and predictive and classificatory purposes" (correlation 0.97÷8.18E-01; p-value 7.67E-55÷1.28E-22) and the second, Dim2 "not-supervised methods; project, simulation, literature, text data; monitoring, decision-making processes; machinery and environment" (corr. 0.84÷0.47; p-value 5.79E-25÷3.59E-06). We answered the research question regarding which method, among supervised, unsupervised, or other, is most suitable for application to data in the construction industry. Conclusions: The meta-analysis provided an overall estimate of the better effectiveness of supervised methods (Odds Ratio = 0.71, Confidence Interval 0.53÷0.96) compared to not-supervised methods.