Conceptual Model for Predicting Safety Behaviour of Construction Workers: A Neural Network Approach

FUTURE WORKPLACES
23 May 2019
11:00 - 11:15
ICC SYDNEY

Conceptual Model for Predicting Safety Behaviour of Construction Workers: A Neural Network Approach

Safety behaviour of workers significantly influences safety performance of a construction project. It is therefore essential to explore methods to predict the timely status of safety behaviour. As such, data mined from safety evaluations could be modeled to give a more accurate prediction of workers safety behaviour on site. Prediction of safety behaviour of workers is a complex construction management problem that involves several parameters.

The paper discusses the approach of neural network and its application in predicting construction workers safety behaviour. The study developed a conceptual model based on the behavioural theory, organisational climate theory, and computational learning theory. As a conceptual model developed with a computational learning theory intuition, the model overcomes the limitations of statistical conceptual models such as unknown interactions and restrictions in hypotheses setting. The model serves as the theoretical basis for construction organisations to monitor, evaluate, predict, and encourage compliance with current processes and design better interventions.