Reactive to Proactive RCA: 3 Strategies to Predict Human Performance
Root cause analysis by its very nature is reactive, and don’t get me wrong, reactive root cause analysis does produce significant benefits. However, if you want a more efficient allocation of resources towards proactively mitigating human error, improving safety, and increasing productivity, you need to think about how you can predict human performance. Things don’t get better simply because we produced an investigation report.
In many of the industries we serve (oil & gas, energy, refining & chemicals, and more), system performance relies on human performance, and mistakes result in costly and devastating accidents. The move from reactive to proactive methods creates a high-performance culture where less time is devoted to figuring out where things went wrong.
How do you predict human performance and move beyond reactive root cause analysis to create a high-performing culture? Here are three strategies.
1. Trend your data using the right technique.
You can predict the future by trending. Keep in mind that you need to trend both frequently occurring events and *infrequently* occurring events. A lot of trending techniques out there don’t do that. Dr. Shewhart developed some fairly simple trending techniques for trending that are not difficult to learn called XmR or Process Behavior Charts. The chart calculates both the way a system varies and limits that, if exceeded, indicate you have a problem. Do you have the TapRooT® Root Cause Analysis for Major Investigations book? Read Chapter 5 to learn how to use Process Behavior Charts in Chapter 5. Or, consider reading Performance Measures and Trending for Safety, Quality, and Business Management.
2. Investigate precursor incidents and use RCA tools proactively.
A precursor incident is a minor event that could have been a major accident if one or two more safeguards would have failed. This is similar to a “near-miss” but we find that the term “near-miss” isn’t strong enough language to get the attention it deserves. A precursor incident informs you that something bad is going to happen unless you take action. It’s a warning. You can predict future problems by paying attention and learning from them.
3. Tap into human factors research to see how workers make mistakes due to system weaknesses.
Humans have a very limited attention span. Our recent series on Skill, Rule, and Knowledge Models discusses human factors research about that. Sometimes we try to fix these problems by telling the worker to self-check which uses more cognitive resources and steals even more of their attention away from tasks. So, we can predict the future when we are relying on the worker to do the right thing in complex work environments. Do you need to develop checklists so the worker has to do less thinking about the steps, freeing up cognitive resources for other tasks? Do you need to provide additional training so the task becomes “second nature” to the worker? Do you need to redesign the human-machine interface so that workers can easily select the right controls and understand the displays with less opportunity to make a mistake? Learn more about human factors and how people make mistakes when system weaknesses are present.
Which strategy can you implement today to predict human performance?
We hope we have given you some things to think about to improve human performance. You don’t have to be an expert in human factors to get better at it. Understanding some of these principles will help you get started.
What if you could learn techniques to stop human error in just 2 days?
Many people think that our core business is teaching incident investigators how to analyze the errors for root causes after the fact; however, TapRooT® Root Cause Analysis is an investigation system and a methodology, not just a reactive root cause analysis tool.
We are in the business of studying human error and continuously improving the TapRooT® Methodology as new research into the factors about why humans make mistakes emerge so you don’t have to. It’s all incorporated into our investigation software and tools.
Attend our upcoming Stopping Human Error Course in Austin, Texas to learn more about how you can predict human performance for proactive improvement.
