Hot Spots for Vessel-to-Vessel and Vessel-to-Fixed-Object Accidents Along the Great Lakes Seaway

In the past decade, an average of 20.5 vessel accidents per year have been reported along the Great Lakes Seaway (GLS) in Canada and the United States, with the vast majority clustered at specific unsafe locations or sites along the route. Sites with an unacceptably high potential for accidents are referred to as hot spots, and these hot spots are prime candidates for safety intervention. Because of the random nature of accidents, the identification of hot spots must be based on robust site-specific prediction models of accident expectation. This paper presents an empirical Bayes prediction model developed for the GLS that considers four types of accident scenarios: vessel to vessel and vessel to fixed objects, for river and for canal or lock sections. Hot spot sites are determined with two risk tolerance thresholds: 95th percentile exceedance (high-risk sites) and 85th percentile exceedance (moderate-risk and high-risk sites). For the 95th percentile threshold and vessel-to-vessel accidents, five hot spots were identified on the 1,600-km length of the GLS studied (excluding lake or port areas). Of the designated hot spot sections, 10 km (60.6%) were located along natural river courses and the rest at canals or locks. For vessel-to-fixed-objects accidents, all high-risk hot spots were at canal or lock sections (15.5 km). Reducing the threshold to the 85th percentile resulted in a 7.8% increase in seaway length that was designated as a hot spot. The locations of these hot spot sections along the GLS were consistent for both thresholds.