Analysis of design-for-safety implementation factors in the Indonesian construction industry: A two-staged SEM-artificial neural network approach

Abstract

Despite the rapid growth of infrastructure development in Indonesia, work safety remains a major concern in construction projects. Design-for-safety (DfS) is a critical strategy to prevent work accidents. The implementation of design-for-safety is one of the most important strategies for preventing work accidents. This study aimed to analyze multiple factors that influence the implementation of work safety design using two statistical approaches: structural equation modeling and an artificial neural network. Structural equation modeling analyzes the relationship pattern between variables and their indicators, and artificial neural network maps various similar patterns to predict variables that influence implementation. Designers, owners, policies, tools/equipment, knowledge, and contract documents positively affect design-for-safety implementation, with design and contract documents being the most significant variables. Accordingly, industry and government agencies are advised to prioritize contract documents and design factors, along with other variables as supporting factors in their programs to accelerate design-for-safety implementation in Indonesia.