The Occupational Safety and Health Administration (OSHA) recommends performing Job Hazard Analysis (JHA) for construction activities to highlight and react to potential hazards. JHA commonly requires having brainstorming sessions to identify steps and associated hazards within construction activities. A company's personnel involved relies on their experience - and often also on the company's internal knowledge represented in the form of safety rules. The complexity and time consuming nature of JHA prevent safety personnel to react quickly to changes in the construction and the schedule. This paper presents a framework aiming to improve access to a company's JHA knowledge. The framework uses ontologies for structuring knowledge about jobs, job steps and hazards. It also includes an ontological reasoning mechanism for identifying safety rules applicable to a given activity. The framework has been tested using a test case. The results of this test case are discussed in this paper and conclusions for future research are drawn.