Manjula, NHCDe Silva, NSandanayake, YGRamachandra, TGunatilake, S2022-03-122022-03-122017-06Manjula, N.H.C., & De Silva, N. (2017). Predicting unsafe behaviour of construction workers. In Y.G. Sandanayake, T. Ramachandra & S. Gunatilake (Eds.), What’s new and what’s next in the built environment sustainability agenda? (pp. 326-336). Ceylon Institute of Builders. https://ciobwcs.com/downloads/WCS2017-Proceedings.pdfhttp://dl.lib.uom.lk/handle/123/17305The construction industry is known to be one of the most accident-prone of work sectors around the globe. Although the construction output is less in Sri Lanka, compared to developed countries in general, the magnitude of the accident rate in the construction industry is still significantly high. Most of the occupational accidents are due to the unsafe behaviour of the workers. Thus, studying the people-related factor in safety is an effective way to manage safety at work sites. This is a concept gaining more interest across industry sectors globally, and has the great advantage of needing the involvement of the individual employees. The paper therefore focused to investigate the factors influencing construction workers’ unsafe behaviours and develop a model to predict unsafe behaviours based on those factors. The factors affecting construction workers’ unsafe behaviour were identified through literature survey. Expert interviews were carried out to validate and generalize the factors found in literature, to the Sri Lankan context. Survey approach was used to collect data and the processed data were used to develop and train an Artificial Neural Network (ANN) model to predict unsafe behaviour of a construction worker. Then training and validation of the developed model under 7 design parameters was carried out using the data on influential factors of unsafe behaviour of 284 construction workers of C1 Building Construction sector. The data were applied to the backpropagation algorithm to attain the optimal ANN Architectures. The findings depict that the success of an ANN is very sensitive to parameters selected in the training process gaining good generalization capabilities in validation session. The model can be used to determine the unsafe behaviour level of construction workers and their safety training needs.enArtificial neural networksConstruction industryUnsafe BehaviourConference-Full-textArchitectureDepartment of Building Economics20176th World Construction Symposium 2017Colombopp. 326-336What’s new and what’s next in the built environment sustainability agenda?chathuri9m@gmail.com