How ExxonMobil debunked and then modernized the Safety Triangle

by | Jan 8, 2017 | Risk Management

You may not be able to reduce the number of major injuries and fatalities by reducing the number of minor injuries. Keep reading to find out what scientific research says about safety triangle and how ExxonMobil modernized it.

The (in)famous safety triangle indicates that:

  1. There is a stable ratio between minor and major injuries
  2. That minor and major injuries have the same causes
  3. That the likelihood of major injuries and fatalities can be reduced by reducing the number of minor injuries

Brief Summary of Research

The reports prepared for the Policy Research Programme of the Department of Health in the UK [1], [2]  concluded that there was no stable ratio for minor and major injuries.

U.S. Bureau of Labor Statistics data from 1992-2007 [3] indicate that work-related minor injuries across many sectors of the economy have decreased in recent decades, but that fatality rates have not been decreasing at the same pace. This further indicates that the ratio between the minor and major injuries is not stable. In other words, decreasing the number of minor injuries will now lead to a decrease in major injuries.

In fact, a study published in 2013 in the American Journal of Industrial Medicine [4] showed that in the US construction sector, states with high fatal injuries had low rates of  minor (nonfatal) injuries. If the Heinrich’s triangle theory was right, we should see the same rate for both minor and major injuries.

Researchers from the UK Centre for Transport Studies [5] used data provided by the UK Air Accident Investigation Branch from 1997 – 2010 to explore whether Heinrich’s triangle assumptions apply to the Offshore Helicopter Transportation business. They concluded that the “potential for sudden failures during offshore helicopter operations, is in a clear contradiction to classic incident–accident pyramidal relationships” (p.114).

Chevron, the oil and gas company, in their study [6] concluded that the Heinrich’s model fails to recognize that the patterns of high-consequence, low frequency mechanisms of injury do not follow the same pattern as low consequence, high frequency mechanisms of injury” p.1.

Additionally, the Behavioral Science Technology (BST), a safety management consulting firm, conducted a study in partnership with seven major companies, including ExxonMobil, Shell, or Maersk [7]. They concluded that serious injuries have different underlying causes compared to minor ones.

How ExxonMobil Modernized the Safety Triangle

In a separate publication [8] Exxon confirmed that the reduction of injuries at the bottom of the triangle does not correspond to the reduction of the major injuries and fatalities. BECAUSE not all incidents have the potential to turn into more severe ones.

In fact, they found that only 20% of all minor injuries can result in major injuries and fatalities.

Furthermore, the setting in which someone is more likely to witness a major injury or fatality is not as common. For instance, these types of injuries primarily occur to those e.g. working at height, lifting operations, and more, which are managed by a set of controls, such as lock-out/tag-out, confined space entry, energy isolation, etc.

In response to these findings Exxon changed the way they classify and investigate incidents. Whenever an accident occurs Exxon will:

  1. Determine the actual severity (near miss, first aid, recordable injury, fatality)
  2. Determine the potential severity, i.e. What could have happened under slightly different circumstances. Could it have resulted in a recordable injury or fatality?
  3. Deploy expert investigators to deeply explore the factors which contributed to a high actual or potential severity type incident
  4. Apply a simple and less resource intensive investigation for an incident with low actual or potential. For example, they could choose to implement the 5WHY methodology.

Click Here to learn more about Exxon’s approach


[1]         K. Taxis, S. Gallivan, B. Franklin, and N. Barber, “Can the Heinrich ratio be used to predict harm from medication errors? Report to the Patient Safety Research Programme (Policy Research Programme of the Department of Health),” London, 2005.[2]         S. Gallivan, K. Taxis, B. Franklin, and N. Barber, “Is the Principle of a Stable Heinrich Ratio a Myth?,” Drug Saf., vol. 31, no. 8, pp. 637–642, 2008.[3]         BLS, “A Declining Rate of Fatal Work Injuries,” Bureau of Labor Statistics, 2009. [Online]. Available: [Accessed: 07-Jan-2017].[4]         J. Mendeloff and R. Burns, “States with low non-fatal injury rates have high fatality rates and vice-versa,” Am. J. Ind. Med., vol. 56, no. 5, pp. 509–519, May 2013.[5]         F. Nascimento, A. Majumdar, and W. Ochieng, “Investigating the Truth of Heinrich’s Pyramid in Offshore Helicopter Transportation,” Transp. Res. Rec. J. Transp. Res. Board, vol. 2336, pp. 105–116, Dec. 2013.[6]         J. Teakle, T. Lughermo, T. mannion, and R. Biddle, “Stepping Out of the Triangle and into the Field,” in International Conference on Health, Safety and Environment in Oil and Gas Exploration and Production, 2012.[7]         L. Johnson, “Is safety pyramid a myth? Study suggests new approach to injury prevention | Canadian Occupational Safety,” 2011. [Online]. Available: [Accessed: 07-Jan-2017].[8]         R. M. Smith and M. L. Jones, “A Hurt-Based Approach to Safety,” in SPE Americas E&P Health, Safety, Security and Environmental Conference, 2013.

What would make this post even more helpful to you?

12 + 12 =

The problem with the HSE 5 Steps to Risk Assessment and how to improve hazard recognition by 31%

You may be missing even up to 50% of hazards if you use simple risk assessment methods. Read on to find out how to identify over 30% more hazards.

A simple, yet powerful technique to improve the effectiveness of your safety training

If safety training in your company focuses mostly on hazards and things to avoid, then the training participants may end up with inaccurate understanding of the risk. You can further improve your training by using the technique described below.

A new personality trait that can predict safe/unsafe behaviours of individual workers

In 1994, researchers from the University of Missouri, proposed a new personality trait called Consideration of Future Consequences (CFC), which was later found to actually predict unsafe behaviours. Read more to learn how it could be used in your business.

How ExxonMobil debunked and then modernized the Safety Triangle

You will not be able to reduce the number of major injuries and fatalities by reducing the number of minor injuries. Keep reading to find out what scientific research says about safety triangle and how ExxonMobil modernized it.

12 supervisory behaviors that will reduce your injury rate and improve safety culture

Supervisory behaviours are directly linked to workers’ involvement in accidents. Read more to learn about 12, most powerful safety leadership behaviours.