Slashdot Mirror
GM Names and Fires Engineers Involved In Faulty Ignition Switch
An anonymous reader writes 'Thirteen people have died because of faulty ignition switches in General Motors vehicles. The company has recalled 2.6 million cars, paid a $35 million fine, and set up a fund to compensate the victims. Now, an internal investigation into the incident has shown that the company was aware of the problem since 2002. 15 employees have been fired over what CEO Mary Barra calls "misconduct and incompetence." The report singles out Ray DeGiorgio, an engineer who allegedly approved a part that did not meet specifications and misled coworkers who were investigating complaints. "He actually changed the ignition switch to solve the problem in later model years of the Cobalt, but failed to document it, told no one, and claimed to remember nothing about the change."
"There's no evidence anyone else knew the switch was out-of-spec at the time, the report says; neither did DeGiorgio tell anyone when issues with the part were brought to his attention multiple times. When one engineer specifically asked DeGiorgio in 2004 whether the switch met torque specifications, DeGiorgio didn't respond. Evidence the investigators gathered showed that he started two e-mails but never sent them. ... Instead, DeGiorgio was consumed by a problem in which cars with the switch were failing to start in cold weather, something the report says was "a personal embarrassment to DeGiorgio.'"'
"There's no evidence anyone else knew the switch was out-of-spec at the time, the report says; neither did DeGiorgio tell anyone when issues with the part were brought to his attention multiple times. When one engineer specifically asked DeGiorgio in 2004 whether the switch met torque specifications, DeGiorgio didn't respond. Evidence the investigators gathered showed that he started two e-mails but never sent them. ... Instead, DeGiorgio was consumed by a problem in which cars with the switch were failing to start in cold weather, something the report says was "a personal embarrassment to DeGiorgio.'"'
This may not be a conspiracy, but it is an indication of a systemic, cultural failure endemic to the company.
While I've seen some engineers do bad things because they were afraid of management, I've never seen a situation in a company this size where the organization was good but one bad engineer was able to release something terrible with no oversight. This is almost by definition of what it means to be a good organization: you shoudl not place tens or hundreds of millions of dollars of responsibility onto your wage slave, no matter how senior he is (never mind that real physical injury may be involved).
It's always, always been bad management, frequently that went straight to the top. But then with most American car dealers we already know that. I find it amusing that they blame the unions all the time, but my two "Japanese" cars, both manufactured in America, have been excellent and are still running flawlessly 9 years later, while my two "American" cars (made in Mexico) I was happy to be rid of at 5 years.
Of course.
I'm somewhat surprised that the company named names. I suppose the result of the investigation made it clear that his intention was only to cover his own ass, which must have tipped the scales.
Now if only we could get names of lawbreakers out of government agencies. I know it will be a cold day in Hell before that happens, but it would be nice
"Here Lies Philip J. Fry, named for his uncle, to carry on his spirit"
According to this NPR story:
http://www.npr.org/2014/03/31/...
Scott Oldham of Edmunds.com had a test drive of the Cobalt in 2004, with a GM engineer in the car. Multiple times Oldham's knee hit the key fob and car shut down.
Also, a major factor preventing identification of the ignition switch issue (or at least providing plausible deniability) is the part number. GM had 2 sets of cars: one set supposedly had this issue, the other did not. Both had the same ignition switch, so if there was a difference between the two sets, the ignition switch was not it.
Now we know the ignition switch was changed, but the part number stayed the same, making it difficult to correctly identify the issue. We're supposed to believe a single engineer was responsible for changing a part but not the part number?
Not that it matters much to me. My car searches start with Consumer Reports reviews and reliability ratings, and so no GM car has been in consideration for a while.
As expected, the report exonerated the CEO, executives who report directly to her and the company's board of directors. Fifteen employees have been dismissed from GM because of misconduct or failure to respond properly as evidence of the ignition switch's defects mounted, Ms. Barra said. More than half of those officials were executives, and Ms. Barra said five other GM employees have been disciplined but remain with the company. Ms. Barra wouldn't identify the employees by name, except to confirm that two low-ranking engineers involved with the design of the defective switch were dismissed. Also fired were lawyers and officials responsible for safety and dealings with regulators, according to people familiar with the matter.
This posting is provided 'AS IS' without warranty of any kind, implied or otherwise.
Why did GM write into their bail-out a few years ago the clause that they cannot be held responsible for malfeasance which occurred prior to that bail out?
Makes me sick thinking about it.
GM's "bailout" was actually a managed bankruptcy with the terms pre-arranged, and bankruptcy in most US states incldues the discharge of liability, not just debts. It is done that way so creditors can't short-circuit the bankruptcy system and just "Wait to sue" until after you're out of bankruptcy protection.
This liability discharge is one of the main features of bankruptcy. It is why the company that polluted the Elk River in West Virginia (leaving the 2/3 of the state without safe drinking water--some of them to this day) declared bankruptcy in short order after the incident--they knew they had no possible defense against the legal onslaught that was coming, and their executives (who were owed sizable bonuses--coal executives really rake it in) wanted to make sure they filed for bankruptcy BEFORE anybody filed suit, because if a suit was pending when they filed bankruptcy that party could go to court to stop bonuses and incentive pay owed to executives from being payed out. Because if the company was facing a bankruptcy judge and had an already-filed suit for billions in damages he would never (EVER) approve bonus payments to executives and would probably listen pretty favorably to a creditor who insisted the executives not be able to loot the place ahead of their judgement.
Who did what now?
It took quite a lot of time, but the NYT posted the report and I downloaded it and read all the report up to the point it makes recommendations about reorganizing some of GM's administrative structure, which I skimmed. The folks involved in this debacle behaved like they were in a Marx Brothers movie. There's the GM Nod in which committee members all nodded that things would be done and when they left the room did nothing and the the crossed arms pointing which meant the individuals crossed arms pointing to others meaning they weren't going to do anything. There seemed to be hundreds of instances when folks couldn't remember what went on in the multiple meetings about the ignition switch issue. There apparently is an urban legend at GM that became standard operating procedure that notes were not to be taken at meetings as well as minutes. No wonder no one remembered what they were told or said. What's it called, probable deniability?
Just one situation out of many struck me as showing the engineers' incompetence: At one point it became clear that model year Cobalts after 2007 did not have the problem with the ignition switch where it would move from run to accessory just by brushing the key fob hanging from the inserted key with clothing. A couple of guys, including an intern, went to a junk yard to examine a car that had been involved in some kind of accident. The intern noticed that the ignition switched required very little torque to switch from run to accessory so the group got a fisherman's scale to measure the torque. They then got appropriate torque meters (Snap-on tools has nice ones which I have used) but only looked at the newer cars because they couldn't find any older ones to test. DeGiorgio had asserted there was no change in the switch torque from the initial design, so I'm guessing they just ignored the junk car result. My guess is they could have looked for old cars at used car dealers or car auction lots for testing or even got hold of the Michigan state motor vehicle department to find owners of older Cobalts. GM should also have a database of Cobalt VINs connected to registered owners. And of course, the ultimate incompetence was that no connection was ever made that when an ignition switch moved from run to accessory mode the air bag sensors were disabled and would have solved the mystery of why air bags did not deploy during accidents when the switch was turned.
This is a very interesting, fascinating and engrossing report and I encourage people to read it. I wonder if it might become required reading for discussion in engineering and law schools.
In a time of universal deceit, telling the truth is a revolutionary act. George Orwell
Because the old GM is gone. The shareholders and management switched. It's a new company with the same name and it doesn't deserve to be liable for the past company.
"Doesn't deserve"? Gotta disagree with you there. Sure the company technically is incorporated as a "new" company and some (but not even close to all) of the management has changed but fundamentally it is still the same company. You are giving them a pass based on some legal technicalities which they do not deserve. In all practical terms it is the same company, selling the same products, under the same name, with mostly the same employees and the same facilities.
I run a company that supplies parts to GM. (we're a Tier 3 supplier) I honestly doubt there was much if any cover up. Frankly in my experience GM is too incompetent for that. I see their engineers do stuff all the time that is borderline retarded and the company is so large it's hard to even find a person responsible for a specific issue, much less hold them accountable. While I can't say for certain either way, I tend to think the cause of this fiasco is more structural than criminal. I think this is probably a case of incompetence of such a degree that it appears as malfeasance.
If he was a licensed PE he had a professional and legal obligation to intervene with the switch, regardless of how he felt about it. If he wasn't a PE, then whomever the PE was that was managing him and approving his designs is to blame.
In automotive engineering PEs are a rarity. There is no requirement whatsoever that a PE be involved or that one signs off on any designs. You find PEs in civil engineering and some aerospace and a few other fields but most engineering does not require such a certification. There would be a production part approval and there would be an engineer of some sort who would be responsible for the design and production. Most parts in US automotive production require a PPAP document to be completed for both design and production processes. It's usually a pointless waste of time but there is a formality to the process and it does assign responsibilities.
a) There was no change management?
b) A single engineer can replace a critical component without anyone ever needing to sign off?
c) Not answering an e-mail does not make one culpable, it merely points to a time management problem or not enough time to respond
d) Even when an e-mail did not get answered, nobody cared enough to follow up?
These things point to serious managerial issues. Engineers can make mistakes, covering them up and pointing the finger is a managerial issue.
Custom electronics and digital signage for your business: www.evcircuits.com
"More than half of those officials were executives, and Ms. Barra said five other GM employees have been disciplined but remain with the company. Ms. Barra wouldn't identify the employees by name, except to confirm that two low-ranking engineers involved with the design of the defective switch were dismissed. Also fired were lawyers and officials responsible for safety and dealings with regulators"
Do you know what you're talking about?
Should the CEO be signing off on every single part that goes into every one of their vehicles?
More than one person should be signing off. Certainly it shouldn't have even been possible to later change the design and sneak it into production without even changing the part number.
I've heard NOTHING that leads me to believe me to believe that these cases of "uncontrolled acceleration" were anything of the sort.
Then you must be an all knowing expert. Oh wait...
http://embeddedgurus.com/barr-...
So a tree falls in the forest but you can't hear it, then it must have never been growing there in a first place!
A little background research show lawsuits were filed on Jan 10, one day after the event, Freedom Industries did not file for Chapter 11 bankruptcy until Jan 17. Really it is more about the limited liability of the company stake holders and officers than bankruptcy law that is upsetting.
If you step on the brake it will overcome the accelerator every time no matter how hard you rev the engine.
I have a counterexample:
5-6 years ago, I was driving my wife's 1997 Ford Taurus when the accelerator pedal stuck to the floor. I pressed the break as hard as I could (both feet and as much of my 220 pound weight that I could put on it from a seated position), but we continued to accelerate. Thankfully, I was able to put the car in neutral before we crashed into anything. I coasted to the center turn lane, put on the e-brake, and sat there calming down, with the engine redlining until I shut it off.
I know with 100% certainty that I wasn't pressing the wrong pedal - the accelerator was still stuck to the floor after I got help from a cop to push the car into a parking lot. This was a mechanical issue (not many manufacturers were doing drive-by-wire throttle back in 1997); the engine had just been rebuilt, and the shop must have reinstalled the cable incorrectly - among other things they screwed up.
This car was fairly old (probably 130k miles at that point), but the brakes were well-maintained, and they were four-wheel disc.
You might be right for some - perhaps most - instances, but not 100%, as my experience proves.
What the media fails to mention is that nearly half of the fatalities related to this ignition switch problem also involved some combination of alcohol, drugs, and lack of seat belt use. Please see the latest issue of Car & Driver for more details, I just read the article last night. This is not meant to downplay the engineering/management mistakes that were made but simply to illustrate all the factors involved with the loss of life attributed to this mistake. I also own two Toyota's that only accelerate when I tell them to...
It's still an incredibly small statistic.
Let's put this into perspective: people are afraid of speeding. I've seen advertisements showing skulls and children, talking about how you have a 70% chance of survival after being hit by a car at 25mph, or a 40% chance of death at 40mph (misleading: the statistics are wrong--you actually have a high chance of death or severe injury around 40mph--and the asymptote inflects around 35mph). States raising speed limits always get a huge political battle over all the dangers of driving 75mph on the Interstate, since 60mph is so very safe. We have signs on buses advertising the crackdown on speeding.
Meanwhile, people are getting licensed with as little as 10 hours behind the wheel of a vehicle. The license test here? As prompted, use your signal and turn left. Then parallel park. Then drive to a stop sign, stop, signal, turn right. Two more stop signs, 30 feet to each. Congratulations kid, ya pass.
Most racing schools also have advanced driving classes. These classes usually start with a discussion about vehicle dynamics, then move on to practical experience driving on closed course. Serpentine course to feel how your suspension loads, handles, and fails. Skid pads to practice skid recovery. Minor obstacle courses where you practice searching for, recognizing, predicting, and reacting to hazards. These courses teach you to handle your car in hazard situations, to recognize potential hazards before the situation becomes hazardous, and to react to hazards that come out of nowhere (idiot drivers, kids appearing from behind parked cars, etc.).
These are all things you will encounter repeatedly while driving, but we teach none of this in driver's ed. We don't require it for licensing. We don't even put you on the road to see how you drive in traffic. Can he stop at a stop sign? Then he gets a license. Put anti-lock brakes in the car, he'll be fine. No need to prepare for rain, ice, blown-out tires, pedestrians, children, other bad drivers, or the simple consequence of encountering the limits of your car when actively reacting to any of these things.
74 people in over a decade isn't a lot. That's 7.4 each year. Training these people for to unpredicted hazardous situations would have increased their chances of recovering or minimizing the damage, even as the brakes and steering became stiff. I've shut my engine off and back on due to a stall, in dense traffic, at 40mph; I never considered an engine restart a dangerous situation, but that's just because I've always handed it properly. I see not everyone can.
Of course we should fix these issues. We should prevent unnecessary life-threatening hazards. I simply don't see this particular engineering issue as worthy of so much attention. It's minor, it had very low incidence of harm, and it's readily fixed. We've learned lessons from it. There are much worse things going on right in front of our faces that we're not getting outraged over, and those things are cheap to fix and causing thousands of unnecessary deaths every year.
Support my political activism on Patreon.