No organization wants problems. But some organizations are better at avoiding them than others. Some are significantly better. High Reliability Organizations (HROs) are examples of high-risk operations involving multiple people and multiple decisions that perform at an exceptionally high level. High Reliability Organizations are not in one specific industry, but their approach to risk and mindset are similar. Operating a nuclear aircraft carrier, coordinating commercial flight operations and delivering medication in a hospital all share an extremely low tolerance for error.
High reliability organizations are high quality operations that are relatively error free over long periods of time. That definition summarizes points made by Karlene H. Roberts, a professor at the Haas School of Business at the University of California. She, along with Todd R. LaPorte and Gene I. Rochlin, both political science professors, led the original ‘High-Reliability Organizations’ project at Berkeley in 1984. They studied the complex operations of air traffic control, electric utility grid management and a US Navy aircraft carrier. The lessons of high reliability can be applied within any organization.
Measuring Unreliability
Reliability is a number. It’s a probability. Each task within your operations has a reliability regardless of whether it’s measured. Like any other indicator, measuring it will provide a reference point for improving it. If you don’t measure the reliability, you’ll have no baseline. Are there any tasks that would add value to your organization if their reliability improved? Determining that value can reveal the annual cost (or risk) of unreliability for that task in your operations.
No organization wants problems, but results vary. There is a wide range of reliabilities across industries. Some operations have a 90% success rate. Others are at 99.999% depending on the task. Consider the differences in operations with a defect rate of 1 in 10 versus 1 in 100,000. Both numbers measure the reliability of a task. But one is 10,000 times more reliable than the other. The objective isn’t to reach the same level of reliability everywhere, but to ensure your operations are as reliable as they need to be to meet the overall goals.
What is your organization's risk tolerance? Is it 1 in 10, or 1 in 100,000? Whatever the current reliability is, is what the organization tolerates. But it can be changed. Consider the example of medication delivery. A hospital may dispense 1,000 doses of medication each day. A 99.9% success rate reflects an average of one medication error per day. The table below shows what different medication rates look like for 1,000 doses per day.
| Successful | Unsuccessful | Medication Error Frequency |
| 90% | 10% | 100 errors per day |
| 99% | 1% | 10 errors per day |
| 99.9% | 0.1% | 1 error per day |
| 99.99% | 0.01% | 1 error every 10 days |
| 99.999% | 0.001% | 1 error every 100 days |
| 99.9999% | 0.0001% | 1 error every 1000 days |
It’s important to understand the reliability of a company’s work processes. Does your organization do something that is potentially life threatening, that involves multiple people in different departments, a thousand times per day? The expectation of every patient and family member is the medication will be correct. Zero errors is the goal.
Error reduction is well understood in some sectors of some industries, but it remains an unknown in too many organizations. It’s not done by telling people to “stop making mistakes.” The effective design of medication delivery requires insight from doctors, nurses, pharmacists and technicians. Many industries are orders of magnitude away from the reliability mindset of how medication is delivered in hospitals. For reference, in 2016 there were ~4.45 billion medication prescriptions in the U.S.
When a problem does occur, the sections of the process that contributed to the incident must be dissected. The discussion must be an in-depth review of where and how the process broke down with the people who were involved. There may be unique situations that are addressed with the individual, but solutions that improve the reliability of the task will be implemented within the task.
Highly reliable organizations have highly reliable work processes. But they are not rigid and prescriptive. They must be comprehensive and account for special circumstances. There’s not a one size fits all for a work process. Contingencies need to be captured as part of organizational know-how and included in people’s training. And if specific people make a significant positive impact on operations, then what they’re doing needs to be captured in the work process so that others can learn what to do. Having an incident because someone took a day of vacation reveals a low reliability organization.
If you’d like to learn more about how we help organizations improve the reliability of their work processes, contact our office or send us an email. Any work process or task within your organization that has costly, recurring issues is an obvious place for you to start. We can present a High Reliability Leadership session online or in person to your and your groups, or you can attend one of our High-Reliability Leadership Online Short Courses.