Immersive virtual reality as an empirical research tool: exploring the capability of a machine learning model for predicting construction workers’ safety behaviour

Abstract

In recent years, research has found that people have stable predispositions to engage in certain behavioural patterns to work safely or unsafely, which vary among individuals as a function of their personality features. In this regard, an innovative machine learning model has been recently developed to predict workers’ behavioural tendency based on personality factors. This paper presents an empirical evaluation of the model’s prediction performance (i.e. the degree to which the model can generate similar results compared to reality) to address the issue of the model’s usability before it is implemented in real situations. As virtual reality allows a good grip on fidelity resembling real-world situations, it can stimulate more natural behaviour responses from participants to increase ecological validity of experimental results. Thus, we implemented a virtual reality experimentation environment to assess workers’ safety behaviour. The model’s prediction capability was then evaluated by comparing the model prediction results and workers’ safety behaviour as assessed in virtual reality. The comparison results showed that the model predictions on two dimensions of workers’ safety behaviour (i.e. task and contextual performance) were in good agreement with the virtual reality experimental results, with Spearman correlation coefficients of 79.7% and 87.8%, respectively. The machine learning model thus proved to have good prediction capability, which allows the model to help identify vulnerable workers who are prone to undertake unsafe behaviours. The findings also suggest that virtual reality is a promising method for measuring workers’ safety behaviour as it can provide a realistic and safe environment for experimentation.

Appendices

Appendix A: The work-specific big five inventory

Please indicate the extent to which you agree with the following statements (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

I see myself as someone who…
Q1: is talkative at worka	Q18: is generally trusting at workb
Q2: tends to find fault with others at workb	Q19: tends to be lazy at workc
Q3: does a thorough job at workc	Q20: is emotionally stable at workd
Q4: is depressed at workd	Q21: has an assertive personality at worka
Q5: is reserved at worka	Q22: can be cold and aloof at workb
Q6: is helpful and unselfish with others at workb	Q23: perseveres until the task is finished at workc
Q7: can be somewhat careless at workc	Q24: can be moody at workd
Q8: handles stress well at workd	Q25: is sometimes shy at worka
Q9: is full of energy at worka	Q26: is considerate and kind to almost everyone at work. (b)
Q10: starts quarrels with others at workb	Q27: does things efficiently at workc
Q11: is a reliable worker at workc	Q28: remains calm in tense situations at workd
Q12: can be tense at workd	Q29: is outgoing at worka
Q13: generates a lot of enthusiasm at worka	Q30: is sometimes rude to others at workb
Q14: has a forgiving nature at workb	Q31: makes plans and follows through with them at workc
Q15: tends to be disorganised at workc	Q32: gets nervous easily at workd
Q16: worries a lot at workd	Q33: likes to cooperate with others at workb
Q17: tends to be quiet at worka	Q34: is easily distracted at workc

1. ^aExtraversion; ^bAgreeableness; ^cConscientiousness; ^dNeuroticism

Appendix B: The safety behaviour scale

Please indicate the extent to which you agree with the following statements (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).

I see myself as someone who…
Q1: overlooks safety procedures in order to get my job done more quicklya
Q2: follows all safety procedures regardless of the situation I am ina
Q3: handles all situations cautiously to not take shortcut as if there is a possibility of having an accidenta
Q4: uses safety equipment required by safety practicesa
Q5: keeps workplace cleanb
Q6: helps co-workers to be safeb
Q7: follows safety practices to keep my work equipment in safe working conditiona
Q8: takes shortcuts to safe working behaviours in order to get the job done fastera
Q9: does not follow safety practices that I think are unnecessarya
Q10: reports safety problems to my supervisor when I see safety problems performed by co-workersb
Q11: corrects co-workers’ unsafe acts to ensure accidents will not occurb

1. ^aTask performance; ^bContextual performance

Appendix C: Machine learning model weights and biases

See Table

Table 4 Weights and Biases

4.