Updated: Sep 15
All regulated industries struggle (to varying degrees) with meeting the intent of regulations that call for the identification of the causes of serious problems i.e. root cause analysis (RCA), and effective CAPA programs. Among these industries, the commercial nuclear power industry within the Energy sector has leaped ahead in causal analysis and CAPA effectiveness. There is a very good reason for that; their ability to learn from their mistakes.
In the past 50 years, the commercial nuclear power industry has been rocked by a number of disasters, each of which threatened to shut down the entire fleet of nuclear plants around the world. These include:
• The Brown’s Ferry Fire (March 22, 1975)
• The Three Mile Island Core Meltdown (March 28, 1979)
• The Davis-Bessy Reactor Vessel Head Degradation (February 16, 2002)
• The Chernobyl Disaster (April 26, 1986)
• Core Damage to Multiple Reactors at the Fukushima Daishi Site (March 11, 2011)
These incidents have driven, in large measure, dramatic improvements in programs, processes and procedures in the nuclear industry. Within the US, it is also an industry with two independent regulators; the U.S. Nuclear Regulatory Commission (NRC), and the Institute of Nuclear Power Operations (INPO), which was created after the Three Mile Island incident to self-police the nuclear power plants in the United States.
The over abundance of regulatory oversight and new and enhanced regulations after each of these incidents resulted in major improvements in the design, fabrication, operations and maintenance of nuclear plants. A key indicator of the U.S. nuclear industry’s success in learning from these accidents and other incidents are the nuclear power plant Capacity Factors (CF). CF is the ratio of actual energy produced by a power plant in a given period, to the hypothetical maximum possible (i.e. energy produced from continuous operation at full rated power).
As causal analysis, CAPA and other programs improved, so too have the nuclear plant CFs . In 1980, at the time of the Three Mile Island incident, the U.S. Nuclear CF national average was 55%. By 2020, the national average CF had risen to 92.5%, a huge vote for the real benefits that can be gained by establishing a learning and continuous improvement culture.
More stringent regulations and oversight have raised the bar for many programs and processes, including RCA. As an RCA subject matter expert and practitioner since 1989, I have combined lessons learned from almost three decades as a practitioner in commercial nuclear power and the Nuclear Regulatory Commission, with the best features of many traditional tools and techniques, into one scalable, lean and agile approach that identifies the deepest-seated causes of complex, human-centric problems in modern socio-technical systems that we work in today. The result is a methodology this is highly scalable, accurate, repeatable, and most of all, efficient. Traditional methods that have been in use (in some cases since the 1940s) were created in another era under different working conditions.
I'd like to illustrate how three traditional tools that are still widely used were not built to simultaneously tackle complex, human centric problems that are a mix of (1) human factors and at-risk behaviors, (2) the human interface with the technology around them, and (3) the latent organizational and programmatic weaknesses that are prevalent in our work environments. These three tools are the Fishbone, the Five Whys and the Events & Causal Factors Chart.
The Fishbone was introduced in the 1940's by Dr. Kaoru Ishikawa. I'm a big fan of his work, as Dr. Ishikawa put cause and effect analysis on the map in a very visual way. The bones in the Fishbone (i.e. man, method, machine, materials, environment) happen to strongly correlate with the main areas in the Anatomy of an Event (shown in the chart below), which we have studied in the nuclear power sector to make sure we cover all the bases when investigating issues. We take the approach that there are 100s of latent weaknesses in our organizations that are causing problems, and every incident or event is an opportunity to uncover and address as many of those latent weaknesses as we can. To do so, our methodology must use every bit of evidence we can gather and analyze, and then to explore relevant information in those key areas highlighted in the anatomy of an event (which we call themes).
However, the Fishbone should not be used for more than the simplest of issues, and certainly not for conducting RCA. The true value of the Fishbone is that it prompts us to evaluate a problem from different perspectives (the bones), providing for a more thorough investigation and a better overall understanding of the causes of the event.
I've also noted that the Fishbone is not taught and used properly in many cases. To be used properly, each branch that comes off the bones of the Fishbone should be treated much the same was a Five Why staircase, but with no limitation on five (a misconception addressed later). Many organizations do not use the Five Whys in combination with the Fishbone for maximum effectiveness. The result is a diagram like the one shown below that is difficult to interpret and unlikely to identify any root causes.
If you consider that each branch that comes off of one of the bones as an individual Line of Inquiry, it would only take about a dozen lines of inquiry to overwhelm the Fishbone and our ability to interpret the results. Here is the last Fishbone I ever did in 2007 (because I was instructed to use the method by a client). Note that even after incorporating the Why staircases and color-coding the common causes, I would still have to explain which ones are considered root causes and which ones are contributing factors.
Let's use the next diagram below of a BlueDragon HCA as an example. In lieu of bones on a Fishbone, we use the themes from the Anatomy of an Event. But the more ergonomic structure of the BlueDragon HCA framework allows us greater flexibility to capture the cause-and-effect analysis. For a RCA to demonstrate the appropriate level of rigor, we typically develop between 30 and 100 Lines of Inquiry, and that many would be difficult to chart on a Fishbone.
The chart below has over 100 Lines of Inquiry. Each Line of Inquiry is a Five Why staircase. Note that our Lines of Inquiry go far past the 5 questions of the Five Whys, sometimes as much as 20 questions deep. If you attempt to draw this diagram on a Fishbone, it would not be a fun day. We color-code the common causes to make it a visually powerful storyboard and the results are clear; the root causes are going to be at the bottom, at the end of all of our cause-and-effect analysis. We can expand this diagram to accommodate as many Lines of Inquiry as we generate, based on the evidence and the complexity of the event. The Fishbone would become unwieldy after about a dozen Lines of Inquiry.
The Five Whys:
There are a number of common misconceptions about the Five Whys, a term coined by Toyota in the 1970s. Here are three of the biggest misconceptions about the Five Whys.
"The Five Whys is a root cause methodology." Not true. Many organizations list the Five Whys in their RCA tool kit as an approved RCA methodology and this, by far, is the biggest misconception out there. The Five Whys is a method to capture cause and effect analysis and the basic building block for conducting a comprehensive RCA. A credible RCA requires many more than one or a few Five Whys because a single (or even a few) Five Whys do not provide any assurance that we will identify the deepest-seated causes. Even if there is a chance that a single Five Why staircase can identify a root cause, there would not be enough "demonstrable rigor" to provide the necessary confidence to ourselves, our managers and our regulators, that we actually identified the root causes. I have reviewed dozens of approved RCA reports that used a single or a couple of Five Why staircases to complete the analysis, and they fall short of coming close to the deepest-seated causes. BlueDragon HCA uses between 30 and 100 Five Whys (we call them Lines of Inquiry) to complete a comprehensive RCA. The Lines of Inquiry are quickly identified when we place available information on the BlueDragon framework shown below.
"We only have to go about 5 questions deep before getting to the root causes." Not true. We continue to ask questions until there are no more answers. And if we went too far (i.e., way past the control of the organization, such as the Earth's gravitational pull, or the Congressional Budget), we go back to the previous, actionable answer.
"Repeatedly asking the question "WHY" is an effective way to conduct cause and effect analysis." Not True. Asking WHY is the most elementary way to conduct cause and effect analysis. BlueDragon HCA uses Socratic Questions, a more sophisticated approach that helps us formulate questions that seek clarification, probe assumptions, probe the evidence and challenge points of view.
Events and Causal Factors Charts:
E&CF charts have also been used for decades and are an improvement over the Five Whys because it includes a timeline, and pulls Five Why staircases (cause and effect analysis) from the timeline. In effect, it greatly increased the number of Five Whys that are used in the analysis. I used E&CF charts effectively, in combination with other tools and techniques, until 2012. The picture below is the last E&CF chart I developed for an incident at a Uranium processing facility.
The strength of the E&CF chart (i.e. the ability to initiate Five Whys or lines of inquiry from the timeline) actually became its weakness. E&CF charts only initiate cause and effect analysis from the timeline, and we need to use other information to develop a better analysis. Information such as the administrative requirements and physical barriers that should have prevented the events (we call them the "defenses"), and the themes from the anatomy of an event to make sure we looked in all the right places.
I spent decades conducting RCAs, audits and assessments where I used the traditional tools and methods, but was frustrated that I had to use a number of separate tools like an E&CF chart and a Barrier Analysis table, which then had to be integrated into a single narrative. The traditional RCA methods were both antiquated and time consuming. We also took subject matter experts from their mission critical tasks for weeks at time, which no one likes to do. Some organizations actually down-classify their events to avoid doing a RCA because they are so time consuming.
Ultimately, I kept the timeline but added two more lines on the HCA chart to capture the themes from the Anatomy of an Event and the applicable administrative requirements and physical barriers that should have prevented the event. And so was the initial stage of developing the modern framework for conducting RCA: the basic BlueDragon HCA framework.
The Basic BlueDragon Hyper-Integrated Causal Analysis Framework
With a timeline, the defenses and the themes in front of us on one chart, we can analyze the information in an integrated manner, dramatically shortening the time it takes to complete the analysis. My average times to complete an RCA has dropped to 5 days. The following tools and techniques are now combined and conducted in a seamless and integrated manner.
Comparative Timeline Analysis
Analysis of Defenses
Themes from the Anatomy of an Event
Ishikawa (bones from the Fishbone)
Lines of Inquiry
Cause and Effect Analysis
Common Cause Analysis
Five Why Staircases and Cause Trees
Events & Causal Factors Charts
Human Performance Evaluations
With the BlueDragon HCA framework, RCAs that usually take weeks and months can be completed in hours and days, while ensuring better accuracy and more demonstrable rigor. Its value is reaffirmed by how rapidly it's being adopted at the nuclear weapons complex and the Department of Energy's National Laboratories such as Savannah River, Sandia, Brookhaven, Los Alamos, Lawrence Livermore and other DOE Uranium processing sites.
RCA is a universal language. And the HCA methodology can be readily implemented in any industry where it is important to conduct a rigorous RCA to identify the deepest-seated causes of problems, and take corrective actions and actions to prevent recurrence (also known as CAPA). These include Aerospace, Healthcare and Pharma, which seem to be lagging in their RCA methodologies.
Tip of the Week:
Shift from the traditional tools such as the Fishbone, Five Whys and E&CF charts to BlueDragon Hyper-Integrated Causal Analysis, a next generation RCA methodology built to identify the deepest-seated causes of complex, human centric problems in modern socio-technical systems.
Ditch the Fishbone but keep the themes, ditch the Events and Causal Factors Chart, and recognize that the Five Whys is merely a building block; the mechanism by which we pursue Lines of Inquiry.
Adopt the lean, agile, scalable, accurate and cost-effective BlueDragon HCA framework. The HCA framework incorporates the important elements of the Fishbone and E&CF charts, and uses the Five Whys as the cause & effect method for pursuing Lines of Inquiry. HCA also dramatically changes the way that RCAs are conducted. For example, we use small teams, and subject matter experts (SMEs) are only used as needed (for an hour or two), allowing them to stay on mission critical tasks. During the analysis, we can bring in over 100 SMEs to participate in a matter of days, working towards the final product from the start. The interview process is also dramatically different: it's totally transparent and we do not take notes - we work on the final chart. The way we conduct HCA also eliminates bias from team members and from those being interviewed. And, the process generates an accurate storyboard that shows demonstrable rigor and makes it easy to defend the cause-and-effect analysis.
Learn to use BlueDragon proactively, using Quick-Hit investigations on the lowest level issues to better identify the causes. These causes can then be coded and input into a trending and analysis program that will identify negative performance trends. Once a trend is identified, a more rigorous BlueDragon HCA can be completed, to identify the deepest-seated causes of the trends. By doing so, we proactively eliminate the root causes that would eventually bypass all of our defenses and cause a major event. This approach is key to establishing a continuous improvement culture.
For more information on how to transition to BlueDragon HCA, contact us at: https://www.dle-services.com/contact-us