PT Notes
Safety Metrics for Process Hazard Analysis (PHA)
PT Notes is a series of topical technical notes on process safety provided periodically by Primatech for your benefit. Please feel free to provide feedback.
This PT Note is the first of two on the topic of PHA metrics. This PT Note addresses Safety Metrics.
Commonly, metrics used for the hazard identification and risk analysis element in process safety programs focus on PHA recommendations, such as the number of recommendations made per study and the number of recommendations unresolved by their due date. Other more detailed types of metrics can be used for PHA studies that delve into the data contained in a study.
PHA studies contain a wealth of information and safety metrics provide a means of generating insights that are not possible through a manual review of study results. Safety metrics provide insights into the origins and nature of risks for a process. They identify when process risks depart from those typical of other processes. The purpose of using such metrics is to monitor and assess the safety of a process in order to focus attention on and drive improvements in safety for the process.
Many different safety metrics are of interest to focus decisions on the most effective ways to reduce risks. For example, the distributions of scenarios according to consequence severity and scenario risk provide an immediate indication of the risk profile for a process that depicts the tolerability of risk. Distributions that are skewed or weighted toward high consequences or high risks are of particular concern. Risk profiles can be compared for different processes at a facility to determine where the risks lie for a facility.
A simple safety metric for nodes is the number of scenarios identified for each node. The more scenarios for a node, typically the more attention that should be paid to managing risks for the node, for example, in ensuring safeguards for the node are robust and reliable. However, that is not the whole picture for nodes. Of even more interest, is the number of high risk scenarios for each node and also the percentage of all scenarios for a node that are high risk. It may be that a node has many scenarios but few of them are high risk while another node with fewer scenarios may have a greater number of high risk scenarios. Similarly, two nodes may have similar numbers of scenarios but one of them may have proportionately more high risk scenarios than the other.
The distribution of scenarios according to their type of cause can be examined for an overall process or for individual nodes. Processes or nodes dominated by human failures merit particular attention because of the generally higher likelihood and greater variability of human failures than other types of causes.
Safety metrics for safeguards can examine the number of safeguards per consequence which can be compared with norms that are typical for a particular company or industry so that deviations from the norms require examination and possible remedial actions. Also, an examination can be made of the balance between different types of safeguards, for example, human versus engineered, passive versus active, and prevention versus mitigation. A large number of human safeguards is problematic because of the unreliability of people, passive safeguards are favored over active ones because of their higher reliability, and it is good practice to have a balance between prevention and mitigation safeguards so as not to be overly dependent on either one.
Safety metrics should not necessarily be used in isolation. Rather, they should be integrated into an overall risk management process in which the sources and nature of risks are examined, the appropriateness of safeguards of different types is scrutinized, and the risks of different processes are compared. Consideration of these aspects of process risks enables more appropriate risk management actions to be taken and assists with resource allocation.
Safety metrics provide powerful ways of processing PHA study results to obtain insights that otherwise are not readily discerned. They support companies committed to data-driven decision-making by providing data-based justifications for informed decisions. The use of PHA metrics helps to maximize the return on a company’s investment in PHA studies.
You may contact Primatech for further information by clicking here.
To comment on this PT Note, click here.
You may be interested in:
Copyright © 2020, Primatech Inc. All rights reserved.
enteiconOf Interest
