The GM Ignition Switch: Good Processes Would Have Prevented This Mess.
The “Ignition Switch” trials are all over the legal news. The choices of the first plaintiffs seem curious. Maybe someone who can tell a consistent story would be a good choice. However, I believe that many of our clients are fretting over how to prevent this type of high profile litigation from engulfing them.
How did a once proud company get it so wrong? What does the backlash against GM’s decision-making (or lack of decision-making) mean to other manufacturers?
I have been stewing about this for some months and I wanted to share my thoughts.
How Did This Mess Happen?
Each of these large, volatile product failures seems to revolve around a theme. The theme that seems to be repeated in the GM ignition switch saga involves the complete lack of communication. There are two sub themes: penny pinching and failure of process. In the end, however, open lines of communication, or a process that enabled communication would have prevented this problem.
We Have Come a Long Way
Over the past 25 years, a whole paradigm has developed for understanding and dealing with risk of injury and the seriousness of hazards. Through the use of formalized FMEAs, FMECAs, FTAs, and similar tools, engineers and designers now spend time formally evaluating hazards, risks, guards, warnings and instructions. I believe that these approaches have resulted in much better design processes and much safer products.
The processes are also far more formalized, which means that they are written down in some fashion for reference and to explain the decisions to later designers and engineers. 25 years ago, engineers and designers were loath to write anything down for fear of being asked about it later. This mindset changed as we got better processes, and grew more confident that the processes and systems yielded better results.
At the same time, I think manufacturers have become far more sophisticated about evaluating and dealing with product issues and failures that do occur. There is a whole industry dedicated to Field Service Bulletins, Letter Campaigns, and Recalls. The way these tools have developed, the urge to “ignore the problem” has given way to a better, more forward looking mindset. Occasionally a manufacturer has turned fixing a problem into a marketing campaign.
Again, I think the current approach has increased safety.
I worry that talk about sending engineers, designers and others to jail will result in a return to the “stone age” of design, where nothing is kept on paper, and design improvements were seen as admitting to a defect. I think that is a logical temptation, but one we must resist.
How Do We Prepare
So what is the best way to minimize risk for the manufacturer?
I have three thoughts:
- Continue to do the things that result in better and safer products;
- Continue to document your processes so they tell the real story of making better and safer products, and;
- Continue to execute your document retention policy to avoid the false narrative.
The reality is that we need to encourage more of what we have been doing. Good processes, properly implemented, yield good results. The better we document how we assessed the safety of a product or a system in 2015, the more easily we can understand and defend the choices we made if we are called to account in 2025.
As far as I can tell, GM did not follow good processes, or it did not let the processes drive decision-making. It is fairly clear that GM engineers knew that there were problems with the switch. Frankly any car buff knows that a heavy key ring will wear out an ignition switch over time.
It is not entirely clear when GM engineers tumbled to the fact that the ignition switch problems would result in airbag shut downs, but they knew it by 2005.
So what happened? I am not sure that we will know the “why.”
How does this change how we should do business? I think it should not change what engineers and designers do, except to reinforce the processes in place and maybe drive us toward better ones.
Good Design Processes Drive Safety
What would have avoided this problem for GM would have been a decision to act based on what the accident data and their own testing told them. Instead, they chose to do nothing for cost reasons.
We should also continue to document our decision-making, because when we do it right, it tells the story of building the safest product. Good documentation is nothing to be afraid of if the design review process is good. If the worst should happen and something does go wrong, we want to be able to say: “we looked at that and we calculated that there was zero risk of that happening.”
Even if we were wrong, a clear, logical documented process results in what I call the “good company defense.” It is much better than:
- “We didn’t think about it”
- “We may have thought about it, but there are no documents”
- We thought about it and maybe it was a risk, but fixing it was too much money.”
Review Accidents and Incidents
And when there is an accident, we go through the process again and let it drive our decision-making. Every accident, incident and product failure gives us new information. We need to use it. It impacts the original calculations on risk and hazard. We simply cannot go through the process and pretend that nothing will change the risk/benefit analysis.
Document Retention Policies: How to Save What You Need
Document retention policies get a lot of scrutiny. The law allows for them, but they are characterized as a “sinister” attempt to hide the truth. I would suggest that the opposite is true. Who hasn’t seen some note or memo in a design, engineering or warranty file that is either misleading or flat out wrong. They range from the memo from the overzealous young engineer who doesn’t really understand the product, to the post it notes scribbled on in the warranty department. These documents are not well thought out and they do not represent the considered opinion of the manufacturer. First, for every memo that proposes an untenable design alternative, there should be a memo explaining why the proposal is a poor choice.
In the end, however, each of these documents should be discarded. Nothing should be kept that does not tell the true story of the design and it’s testing.
In contrast, the outcome of a formal FMEA, FTA or other design review should be kept, as should everything that tells us about the actual design and the reasons behind that design. We want to trumpet what was considered and why some things were rejected. We want to emphasize what safety aspects were studied and what risks were considered.
So, the climate may be a little more hostile to manufacturers. I think ultimately, the hostility will be to manufacturers who appear to ignore problems that in hindsight are obvious. The real question will be: “Were you careful?” And I think we will be required to prove that we were, that we considered the ways the product could be used and the way it would likely be misused.
There will be a push against the “one and done” type of engineering. We will need to reevaluate periodically after an event (a major change of one system or part, an accident, a significant warranty claim). Again, the question will be, “Given what happened, were you careful?”
The best way to defend yourself in the modern climate is the same way it was in the previous climate: Have a truthful narrative that demonstrates that you care about the product and the safety of its users. Your design documents can frame and support that narrative. It is a good idea to examine those processes and recommit to following them rigorously