Abstract
Risk matrices are used in process safety to rate and rank risks of hazardous events to help with decision making on risk reduction for processes. For example, commonly they are used in process hazard analysis to rate the risks of hazard scenarios. Flaws in their theoretical framework and mathematical inconsistencies in their use have been identified in the risk analysis literature. These flaws pose potentially serious problems for the application of risk matrices in process safety. One of the most significant flaws is the potential for risk ranking reversals wherein an incorrect assignment of a higher risk event to a lower risk level and a lower risk event to a higher risk level may occur. This article discusses the nature of the problem and describes how it can be addressed for risk matrices used in process safety.
**
Some comments from the full-text paper:
- “risk matrices. . . may be creating no more than an artificial and even untrustworthy picture of the relative importance of hazards, which may be of little or no benefit to those trying to manage risk effectively and rationally” (p.1-2)
- “One of the most significant problems is the possibility of risk ranking reversals, that is, a higher risk event being assigned to a lower risk level and a lower risk event being assigned to a higher risk level … However, this assumption can be invalid and unrecognized. It is possible for risk matrices to assign the same risk level to quantitatively different risks that should be distinguished. Moreover, a lower qualitative risk level can be assigned to a quantitatively higher risk and vice versa” (p2).
- The author notes that risk matrices on the surface may appear to be simple and useful risk management tools but have pitfalls for those unaware, including mathematical inconsistencies and flaws, which may result in illogical decision requirements.
- For risk reversal, the author suggests correlating risk levels and the decision requirements with iso-risk bands and mapped to cells in the risk matrix, which can help provide logical and conservative results.
- “Risk matrices that conflate risk values because of risk matrix symmetry … do not account for the aversion of people to high severity events … generally people are more averse to an event that causes 10 fatalities every 10 years than events that cause a single fatality every year, even though each has the same risk” (p.5).
- It’s said that matrices can’t provide aggregate measures of risk across a process or facility, and further risk-informed decisions should consider other attributes like cost, feasibility and value in addition. Value trade-offs are said to possibly “complicate choices between risk reduction measures” (p.5).
- The author says that “Risk matrices provide only approximate, not definitive, estimates of risk. Their role should be no more than providing initial decision guidance, which should be used with caution and the application of common sense. Risk matrices should not be used in isolation to make decisions” (p6).
Author: Baybutt, P. (2015). Proc. Safety Prog.
Study link: https://doi.org/10.1002/prs.11768
Link to the LinkedIn article: https://www.linkedin.com/pulse/designing-risk-matrices-avoid-ranking-reversal-errors-ben-hutchinson
SafetyInsights.org 