To Root Or Not To Root: Part 2

Published: Wednesday, November 26, 2008 - 09:41

Part one of “To Root Or Not To Root,”presented a dilemma that many managers face all too often. Regardless of approach, does a problem warrant a full-blown root cause analysis? A scan of the comments to part one indicates that root cause analysis has different meanings to different people. Some readers recommended a full analysis while others thought that would be overkill, preferring a quick fix.

I asked three high-level corrective action gurus in Fortune 500 companies what they would recommend. One expert opinion is printed verbatim below; the others are woven anonymously into the analysis because of respective legal constraints.

Many respondents encouraged taking a look at the bigger picture surrounding this hypothetical scenario. It’s obvious that our fictional company, Kulshan Industries, is experiencing serious growing pains—serious enough that if they don’t make significant and immediate changes, they likely won’t survive in the long term. The company evolved from a business model, which remained virtually unchanged for decades, into a more complex and uncertain environment. This means the company now operates in a world of greater risk. If we zoom out to see the bigger picture, it’s clear that the additional uncertainty experienced by Kulshan is the norm—not the exception. This is increasingly true for all players in today’s aerospace supply chain. Granted, these changes were made to leverage the advantages of specialization and decrease costs. The potential benefits are huge, but to realize the highest potential return from this streamlined system, every player must also completely rethink his or her approach to organizational problem solving. Those who succeed in the medium- and long-term will do so because they have developed an effective problem-solving culture flexible enough to transcend traditional boundaries.

In the following, Gary P. Bosworth, director of engineering for Mission Assurance and Performance Excellence at Raytheon Space and Airborne Systems, gives us his take on our hypothetical scenario:

“There is a great adage about people being penny wise and pound foolish. I see it here. Nancy does not want to tie up labor hours in overhead activity that does not readily contribute to the bottom line. Managers who are very focused on cost containment avoid investing too much time with issues that they optimistically believe can be solved using very simple means.

“Nancy is relying very heavily upon reward and punishment systems to avoid errors from operators. However, she is missing that people do make honest mistakes for a variety of reasons that may or may not be in their control. So the probability of the event reoccurring has not been mitigated significantly.

Knowing that this brand of solder had periodic contamination issues means Nancy is putting the business at risk of future product issues.  Nancy has forgotten to ask herself, “What are the consequence to the business associated with reoccurrence of a quality escape?”  How would the customer react to discovering quality issues that they find in using Kulshan’s product?  What would be the impact to Kulshan’s business if its customer gave it a SCAR (supplier corrective action request) that would potentially affect its supplier rating with that customer? 

When one looks at the potential costs (sales, reputation, etc.) of incurring the risk, the cost/benefit relationship of risk mitigation becomes more readily apparent.
    
“Using a structured methodology such as root cause analysis, with a cross-functional team of people that may have knowledge of key drivers to the problem, is very powerful. By developing backward the cause and effect trees, the problem space becomes defined and the impact of the various branches to the problem can be assessed.

“At Raytheon, when we use root cause analysis, we uncover several contributing factors allowing us to develop solutions that can greatly reduce or eliminate the risk of reoccurrence. Because we recognize people can and do make mistakes even with well-conceived training, we look hard at eliminating conditions or situations (error-proofing the process) to mitigate the risk due to human error.

“In this case, Jodi’s estimate for three full days with five people appears excessive based upon our own experience. A team composed of the right people, that has a strong facilitator, and ready access to key data, can accomplish quite a lot in a short amount of time.”

“If I were Jodi, I would make the case to the vice president that—given the potential impact to the business on reputation, sales, and profit—relying on a carrot and stick system to try and avoid human error will not be the most effective way to detect known material issues. Instead, investing in solving the problem through more robust means not only would mitigate the risk to the business, but could be used to demonstrate a selling point to Kulshan’s customers: an active quality management system focused on improving processes and systems. And as an added bonus, training additional facilitators to be ready to address potential future issues would strengthen Kulshan’s responsiveness to quickly address and solve those problems.”

So let’s analyze the specifics of this hypothetical scenario within the context of Gary’s perspective, as well as the thought leadership from the other corrective action leaders with whom I consulted.

All stakeholders who have an investment, share, or interest in the situation likely agree on the desired outcome—they want to prevent the problem from happening again. But there are many factors that govern the ultimate position any given stakeholder supports.

This can be seen in both Nancy Murray’s and Jodi Ullman’s positions. Nancy wants to conduct a quick, cursory analysis, and then move ahead with a simple solution. Jodi prefers a more thorough approach in the belief that this will provide a more complete understanding, thereby delivering better solutions.

As stakeholders, both have the goal of preventing problem recurrence. However, their positions are based on each manager’s past experiences with root cause analysis. These experiences are more likely to be different than similar. The reason is two-fold. One, there are several ways to conduct a root cause analysis. And two, due to differences in capability and experience, variability exists among users of any given method.

If it were possible to sample from the universe of all root cause analyses conducted over any given year, it’s safe to say results would range from wildly successful to dismal failure.

In light of this understanding, should the vice president sign off on the root cause analysis? My take on the answer, based on my own experience and the majority of reader feedback is absolutely. This should never have been elevated to the vice president in the first place.

Even though the information available is limited, it indicates classic gaps in Kulshan’s problem-solving capability that would be uncovered by a full root cause analysis. These gaps are:

• Stakeholders not identified and notified. The first mistake made by Jodi Ullman and Jerry Smith was to react without first identifying and consulting stakeholders. Had Nancy Murray and other stakeholders been made aware of the problem and its significance before deciding to take action, Jerry and Jodi would have had an easier time obtaining buy-in both to proceed with the root cause analysis as well as to implement solutions.
  
  Stakeholders need to know when a significant event has occurred before additional steps are taken. An ad-hoc notification process indicates that Kulshan doesn’t have the underlying structure in place to support a problem-solving culture. A more mature structure would have a mechanism in place for identifying/notifying stakeholders.
  
• No common goals for root cause analysis program. It’s obvious that Kulshan is lacking a common set of cohesive goals. If Kulshan had goals, there would be no question that this problem represents a significant deviation and therefore warrants a complete root cause analysis.
  
  

  The best root cause analysis programs have specific goals upon which the entire root cause analysis program is built. Because this is such a crucial component, Kulshan should develop simple, achievable goals that facilitate the creation of threshold criteria—specific guidelines that govern when a formal root cause analysis is required. Goals also allow the creation of metrics that provide leadership with visibility into the health of the program at any given time.

• No  common problem-solving method. In this case, Kulshan has done some root cause analysis training, but the extent is largely unclear. It seems clear that the purpose of the training was more to provide “tools for the individual tool box” with the hope that people will select and use the appropriate method at the right time.
  
  

  Individuals solve problems better over time when they use a standard problem-solving methodology. The same is true for organizations. When everyone uses the same method, the organization solves problems more effectively. The organization will be more effective if it commits to a consistent method of root cause analysis for conducting formal investigations.

• Action-based solutions. Nancy comes from a culture where disciplinary action is considered an effective means for controlling behavior. The fact that they had a production meeting about solder contamination issues that resulted in the decision to make operators responsible for managing the associated risks seems ridiculous in today’s advanced quality environment. While it may not be the norm, those of us who make our living investigating serious incidents learn to expect the unexpected—and this scenario is not as far-fetched as some may believe. Unfortunately, the actions or inactions of people are often the primary focus of an investigation and the resulting solutions.
  
  The electronics specialist was set up to fail—virtually the entire risk of contaminated solder was transferred to him and others sharing his role. When a failure does occur, as it inevitably will, it will be the “fault” of the electronics specialist. He will atone by being counseled, trained, cautioned, or maybe even fired. Misuse of discipline occurs less and less these days, but in some industries and some companies it’s still prevalent.

Disciplinary action is rarely an effective long-term solution because it often fails to control the causes of the problem. In the rare occurrence that discipline does work, the effect is limited to the individual—and possibly those who work closely with him or her. Focusing on action causes without finding and fixing the environmental, or conditional causes—those causes in the environment in which people operate—will be ineffective at reducing risk in the system.

In fact, discipline, especially when misapplied, causes more problems than it solves. If discipline is the expected result of an investigation, what incentive is there for those involved to participate? Instead of finding causes, effort is made toward deflecting the blame.

This goes for customer-supplier relationships as well. While many companies have moved beyond the individual blame/punish strategy, the same cannot be said for customer-supplier relationships. Failures often lead to payment concessions, loss of present and future contracts, and other means of holding suppliers accountable. In the current reality of outsourcing, it’s rare that customers and suppliers own problems jointly. This leads to the same problems listed above, only now the tension is between organizations rather than individuals.

• Systemic causes not recognized. Simply defined, systemic causes represent weaknesses in the system that increases the overall risk of failure. Schedule pressure, tension between production and quality, accepting operational risks (such as “watching out for” contaminated solder), and others are all examples of systemic problems. Unless resolved, these systemic issues are likely to be factors in future problems as well.

To most, this may seem like a simple scenario with a straightforward solution. But a full root cause analysis conducted by a competent facilitator would uncover systemic flaws that should cause leadership to wake up in a panic—if they were able to sleep at all.

About The Author

Brian Hughes

Brian Hughes is vice president of Apollo Associated Services -- provider of root cause analysis consulting, training and software. Brian has led significant safety related incident investigations, including those related to major explosions, chemical releases, consumer product contamination and supply chain processes. Brian has helped clients achieve significant savings and improvements in safety, reliability and quality. For more information, visit www.apollorca.com.