Investigating Human Performance in Electrical Incidents
The chapter on an arc flash or electrical shock accident does not end with first aid. While training a class of experienced workers this week, a student discussed a contact incident. After making contact and receiving an electrical shock, the person sent another worker to close the panel. When that worker returned, the injured was found dead. Like arc flash, electrical shock can also result in a delayed fatality.
This incident recall was familiar, since a month earlier, we had investigated an accident that had possible contact with an electrical arc flash. Fortunately, the injured in our investigation is still around, although scarred.
Those can be interesting stories if left at that. However, is it not imperative to dig deeper into why these incidents occurred? Interestingly, the NFPA 70E(R): 2018 Article 110.1(J) Incident Investigations directs organizations to investigate electrical incidents even when no injury or damage to equipment has taken place. Several hundred examples of scientific literature in the form of journal transactions and books require investigators to explore the human portion of the incident.
The Human Performance Modes and Associated Errors
This is fairly simple at larger organizations that employ Human Factors (or Ergonomics) experts, but what do the smaller companies do? They can hire a consultant. However, in many cases, the investigation hardly ever reaches that level of detail – at least from our experience as being third party Electrical Safety Subject Matter Experts.
To assist in promoting knowledge in this area, the NFPA 70E(R):2018 includes an Informative Annex Q Human Performance and Workplace Electrical Safety and requires proactive control via the Written Electrical Safety in NFPA 70E(R):2018 Article 110.1(H)(2) Human Error.
In the NFPA 70E (2018), model human factors are presented in three categories that are paraphrased as follows:
A) If the disconnect switch is operated, the downstream load will be de-energized. Problem: the handle may have moved, but the contacts didn’t separate (Rule-based error).
B) It is compulsory to test before touching any live circuit part. The worker tests the 480V circuit (that is locked out & tagged), measures near zero and proceeds to remove the insulating glove and perform work. Problem: Unaware of the 208V generator that is being plugged into a remote circuit that will back feed through the 480V/208V transformer (Knowledge-based error).
C) The company regularly switches a disconnect as part of an operational routine. As the worker changes to “street” clothing to go home, the foreman requests this switching. Since this is such a routine task, the worker continues to change into their “street” clothing and then switches the disconnect on his way out (Skill-based error – complacency).
Once the principles of the error modes become known, the organization can determine error precursors, e.g., time pressure, personality conflicts, assumptions, etc. and tackle these using existing tools. These tools may consist of a work plan followed by a pre-job briefing, safe operating procedure, three-way communication, stopping when unsure, etc.
These efforts are of paramount importance to prevent an incident from occurring. More importantly, however, these aim at gradually improving or optimizing the workplace culture. Sometimes, as management or field workers, the language we use (verbal or body language) can force a colleague into one the three errors modes described. The tools discussed above take people variability, plant (process) variability, and equipment variability out of the equation. Work can still be performed with a sense of urgency but performed safely.
Just remember, a sense of urgency is NOT the same as an imaginary operational emergency! A safe culture is in no way slow or backwards.
See our previous blogs related to this topic:
Well said!