Human Error – The start, not the end
An incident investigation should not end with human error. That label is used constantly, but it’s too generic. Specifics within an error must be identified. The root cause analysis should contain exactly what the error was. If a valve was mistakenly left open, don’t label it as "human error," label it accurately as "valve left open." Organizations have a bad habit of confusing the classification of an incident with analysis of an incident. Classification and analysis are two separate activities.
Many organizations use an old model for addressing human error. They find out who made the error, then tell them to stop. This can include coaching, counseling or redirecting the employee. It also involves letting the employees know how serious the company is about preventing errors. It may start with a letter in that employee’s personnel file. It may then escalate into a few days off or even termination, depending on the magnitude of the mistake. The message is, “Don’t make errors – we’re serious.” Creating a perception of blame within an organization is the most effective way of disengaging people from the company’s problem-solving effort.
Focus on Process
Disciplinary measures may be a common approach for addressing errors, but it’s not the only way. Disciplining someone for something that was unintentional or unclear doesn’t necessarily make the risk of that error lower. Any corrective actions focused on the individual neglects the bigger opportunity of addressing the work process. Fixing one person, so he or she never makes that same error again is a different approach than fixing the work process, so that no one on any shift, at any company location makes that same error again. There is a difference between an error reduction effort that looks at the individuals versus the work process. People make errors, but the solutions are in work processes.
Initially, management thinks a process focus is soft. The perception is a process approach doesn’t want to hurt anyone’s feelings or hold anyone accountable. Fortunately, it has less to do with feelings and more to do with how exactly to perform the specific steps within a task. That is an objective exercise. The most knowledgeable people on your work processes might be the PhD in the room or it might be the frontline guy who works on 3rd shift. It depends on which part we’re talking about. Focusing on work process will set expectations at a much different level of detail well beyond what most organizations consider adequate. A process approach doesn’t erode accountability, it raises it.
For the incidents we work, the insight and ideas that become part of the more defined work process are provided by the people closest to the work. Initially, management doesn’t understand how the people who make the errors also have information to prevent them. The conventional model is direct employees to do the work. That approach doesn’t necessarily access the best way to accomplish a given task. There are strict technical and compliance requirements within certain tasks, but even inside those steps there is room for innovation. The gap between how management and the frontline see those steps is where errors begin. But it’s also where errors can be prevented.
People ask us, “How do I get started with the ThinkReliability approach?” Fortunately, you don’t need to begin by changing your entire company. You only need to start with one problem. Thoroughly working through one problem will provide a model for addressing other issues in your company. Experiment with our approach in your group with management and your frontline people. Get your own evidence. And don’t hesitate to contact us for assistance.