The construction industry is a high-risk sector and is considered the most hazardous industry worldwide. The impact of construction accidents can be devastating to a country’s macroeconomic condition. This emphasizes the need for effective safety control systems and preventive measures, particularly in identifying safety hazards. The use of association rule mining for construction safety has been increasingly explored in several research. However, there is currently no standard model to assist practitioners in utilizing data mining to identify hazards in construction projects. CRISP-DM, as the de facto standard data mining model, can be implemented to serve as a standard and guide for practitioners. The primary objective of this research is to identify a suitable conceptual framework for construction safety hazard exploration using CRISP-DM. The methods used to develop the framework were literature review and expert judgment. The results of this study indicate that the CRISP-DM model can be applied in the construction industry, but with the support of an improved reporting system and accident data documentation.