From 1999 to 2003, 64,000 accidents and 100 deaths were determined to be attributable to hazardous machines in Québec industries. Although the tools and methods for estimating the risks associated with this equipment offer a first-line prevention strategy, the study revealed that they come in very different forms.
As part of a thematic program intended to provide a better understanding of the processes for estimating the risks associated with industrial machines, the study explored the theoretical efficiency and limitations of a sample of these tools and methods. The researchers hope the results open up avenues for making a better selection of the means of estimating the risks associated with industrial machines best adapted to the hazards to which workers are exposed.
Researchers chose and analyzed the tools in order to characterize the similarities and differences based on equivalence scales for the parameters. This approach was used to analyze different tools using common benchmarks.
The results show that the structure of the tools and terminology used in the tools can potentially lead to biased or incorrect risk estimations. The factors that designers and users of risk estimation tools should consider include:
• The definition of the risk estimation parameters;
• The number of levels or thresholds for each parameter;
• The definition of each level or threshold for each parameter;
• The gaps between levels or thresholds;
• The definition of the exposure interval; and
• The number of risk levels.
Researchers particularly were concerned about the large variety in the terminology or names given to the parameters used for risk estimation, noting, “The most marked differences are found in the definition of the severity of harm (S) and the probability of harm (Ph) parameters. For example, for the S parameter, one finds definitions such as ‘hazard in terms of the potential to cause harm, consequences or potential severity of injury, severity of injury or illness, consequences, severity and severity of harm.’ For the Ph parameter, definitions such as ‘probability of harm, probability of occurrence of harm, probability or likelihood of harm occurring frequency of occurrence, likelihood, likelihood level and qualitative measures of likelihood’ were found.”
The report investigated the differences in results when using different machine safety risk estimation tools applied to the same hazardous situations. The influence of the types of risk estimation parameters used in the tools, the construction or architecture of the tools, the influence of the number of levels for each parameter and the influence of the number of risk levels on the results when applying each tool to hazardous situations were studied. The risk estimation tools were selected based on predefined criteria and compared in estimating the risk level associated with 20 hazardous situations. The results show a large difference between the tools in evaluating the same situation.