Alaska Airlines 737-900 MAX loses a door in-flight out of PDX
Isn’t it surprising that the OEM <and Spirit> people <and IT/paper systems > working on the airplane<s> don’t know:
- the parts they are working on, or
- the procedures for that, or
- intentionally did not comply with them or
- how to assure that everything put together safely
Training may be <is>an issue, discipline <is an issue> to follow approved procedures, and/or workload.
But most certainly management which is responsible for all of it.
How can the FAA not pull the PMA?
- the parts they are working on, or
- the procedures for that, or
- intentionally did not comply with them or
- how to assure that everything put together safely
Training may be <is>an issue, discipline <is an issue> to follow approved procedures, and/or workload.
But most certainly management which is responsible for all of it.
How can the FAA not pull the PMA?
What is very unsettling is that Boeing and parties involved don’t seem to be capable of (quickly - using months instead of days and still no clear answer) finding the required Information. This is by know a “systemic failure” which goes beyond a scope of only Boeing and Spirit and FAA. The ‘door opening plug’ by now has become kind of no more or less than a symptom.
Next to my day to day work at an aerospace OEM I had a special role to periodically solve a Top X list of final assembly and (at)delivery issues. Causal could be any issue - programs, project, line management, design engineering, manufacturing engineering, production engineering, logistics, tooling, purchasing, part and components manufacture, final assembly, flight test, customer rep. I find it very hard to believe that Boeing and Spirit don’t have experienced knowledgeable people in a role like that. People who can quickly come up with a proper answer.
Just thinking …
An aspect that may be ‘underlying causal’ here is the way in which US aerospace manufacturing and production have historically been organised.
A much simplified explanation - work in the US was more cut up in smaller and specialised packages than in for example Europe. You had to adapt work going from Europe to the US and from the US to Europe to this different way of working. The US advantage, if it ran right, was that people could be quickly trained (one example 1 versus 12 months for a specific final assembly worker) and made operational. This initially helped reduce NRC faster. The US disadvantage was that you had mono-functional and sometimes even mono-product people. This put a brake on gaining wider experience and reduces redundancy in the workforce. This suggests that a U.S. operation is less robust in a dynamic operations environment and certainly less robust to shock (… think 787 and MAX programs turmoil …).
If this is your starting position, then squeezing out costs and experience and training and checks removes redundancy and ‘plug’ cases just become symptoms of what could happen in multiple areas.
Obviously the person replacing the door was not the same person as he who removed the door. So how was he to know that bolts were required to lock the fasteners in place... Aircraft have holes drilled in many places for lightness, and without any bolts on site, who is to say if all the holes need to have a bolt. He could have thought that the lower springs where some form of over-centre locking device.
"What is very unsettling is that Boeing and parties involved don’t seem to be capable of (quickly - using months instead of days and still no clear answer) finding the required Information. "
You would think that management could just ask who did this and that person or persons, having not been issued the work instructions for what they did do would stand up and ask for a cigarette and a blindfold.
The basics are basic enough. Spirit needs to stop shipping so many defects to Seattle that there needs to be a full time fix-it response team. Boeing and Spirit need to ensure via training that all employees know not to perform any work that isn't on the work order, particularly when the job requires tools and the detachment or attachment of hardware from/to the aircraft. Finally, as I mentioned above, there needs to be one person with a full time job to search the factory and any time they find white zip/cable ties, to take them and replace them with fluorescent orange ones. With this in place there are probably a few dozen other changes to be made that weren't a part of the door debacle that are likely to prevent some other stupidity from getting out the door.
As an aside - I worked briefly with an efficiency expert, one actually good/useful at his job. He had been brought in to review a new factory. His expertise is fine tuning operations, so when he arrived and they showed him an empty building he said he improves things; he's not a factory planning expert. He was under contract with nothing to do while management tried to figure out how they had f'd this up. Bored, he wandered over to another, operational factory. He was told that the welders were sitting around doing nothing and no weldments were going to the assembly line so he went to investigate.
He said the toughest part of his job is to stop the clams from slamming shut, as many in his field are looking to make the workers work harder. His approach it to make it so their jobs are easier; they still need to be done, and need to be correct, but if obstacles are removed most workers are happy to do the work.
He soon found that these workers were welders and the welders were surrounded by piles metal parts and he asked, just curious right?, why they weren't welding the metal parts together. They told him that they had welded all the pieces together that they could, but all the things around them were missing parts that the machine shop had not supplied. Oh. So he goes to the machine shop and finds that they decided that they would run parts 100 at a time. OK. So until they run all 100 of each part of the welded assembly, the welded assembly could not be finished, so the assembly line had nothing to assemble. But 100 at a time was "efficient." A typical assembly has 10-20 pieces, so the weld shop would end up piled with 1,000 to 2,000 parts before they could complete some items. And that is for each welded assembly, out of 50-100, so the machine shop plan was possibly have 50,000 to 200,000 parts piled around the welders on the last day of machined part production.
Conversations were had. Part runs were smaller. All the parts for each assembly were grouped. The welders said they were tired of getting beat up for not producing and were tired of people asking what about all the parts piled around them. He brokered the deal that they were authorized to sit on their butts for as long as they wanted if they could show that parts were missing. It's a small enough factory that no, they couldn't chuck parts over the fence and yes, the majority of workers were motivated by a job well done. This transformed the job. Whenever parts piled up around the welders, no one was thinking the welders were goofing off. Instead they wanted to find what was needed to get the welders back to welding.
These welding clams were extremely happy, as was the efficiency expert, those in charge of the assembly line, and management. The customer, who had seen the project delayed and delayed and delayed was also happy. And it cost almost nothing besides a few minutes to change over which parts were produced.
I expect something similarly stupid that will cost nearly nothing to fix but will make huge improvements in production rate and decrease error rate can be found at Boeing.
Now for a pun. The problem at Boeing might very well be the Silence of the Clams. To figure out the problems, avoid making the clams close up.
Well the spring-loaded hinges at the bottom, the pin and ramp arrangement at the top pushing the top of the plug away from the fuselage when the springs extended, and the fact that it wouldn't close at all unless something held it in place . . .
