Linked data system for sharing construction safety information

Abstract

Despite the consistent growth of global markets and emergence of cutting-edge technologies, construction jobsites remain among the most dangerous workplaces with extremely high accidents rates. In the safety management process, it is necessary to review and assess a variety of safety documents and data. Numerous public agencies provide open access to accident data, safety checklists, templates and other useful documents for safety management. However, these safety related contents are multifarious, fragmented and often stored in unstructured ways. As a result, the process of finding and utilizing required safety information for specific safety tasks is challenging and inefficient. In order to address this issue, this paper proposes an innovative approach using linked data and semantic web technologies for integrating and sharing construction safety information from diverse sources. A construction safety contents ontology is developed, and construction safety information from accident cases, Job Hazard Analyses (JSA), safety rules and training contents is integrated through RDF formats and SPARQL queries. In order to enable convenient and effective retrieval of information, this study develops a categorization and classification of safety-related contents for safety management processes. The proposed system is expected to improve access and sharing construction safety information, reduce data search time and improve data accuracy. Through this, the approach may enable reductions in safety incidents and overall improvements in safety management and performance.