The rate of recordable injuries in the electric power sector has declined steadily over the past decade; however, the rate of serious injuries and fatalities (SIFs) has plateaued. Unfortunately, studying SIFs is a paradox. On one hand, SIFs are incredibly important and deserve significant resources for investigation. On the other, learning from these events and detecting causal patterns are challenging because SIFs are relatively rare. To vastly increase the number of learning opportunities and to better characterize safety performance, organizations are beginning to investigate incidents with the potential to cause serious injuries or fatalities (PSIF). PSIFs also offer an opportunity for shared learning, which is necessary to advance toward SIF elimination.
Unfortunately, existing methods of identifying and tracking PSIFs are unscientific, heavily biased, and yield inconsistent understanding of whether an incident is a PSIF or not.
An EEI working group of 20 safety leaders and a technical advisor was assembled to create a method for consistently classifying safety incidents and observations that enables shared learning. The EEI safety classification and learning (SCL) model leverages the latest scientific knowledge and the best features of existing methods. The model was tested and refined by the team using actual safety cases.
The resulting tool defines safety incidents and observations based upon the answers to the following yes or no questions:
1. Was high energy present?
2. Did a high–energy incident occur?
3. Was a serious injury sustained?
4. Was a direct control present?
The associated report provides detailed guidance to answer these four questions objectively and consistently. The SCL model is graphically depicted in Figure 3 on page 8. Using this model, consistency in incident classification among the EEI workgroup increased from a baseline of 65 percent to 95 percent. This SCL model enables a common understanding of safety learning opportunities underpinned by a set of definitions for each of the seven incident and observation types in the model. This common language serves as the foundation for shared learning.
Tracking and learning from PSIF could redirect attention from lower–severity incidents to conditions that have the potential to be life–threatening or life–altering, which would be an important step toward the elimination of SIFs. In the future, the SCL model and the associated definitions could be used to form new, more impactful safety metrics that complement traditional indicators like total recordable injury rates (TRIR). This would allow organizations to monitor progress toward the most important goal: saving lives.
CLICK HERE for the full report
