Boeing 737 MAX family NEWS ONLY

[357Mag]

All -

As regards " re-work "...

During my last years @ Spirit Aerosystems Tulsa, QA sent out a message that doumentation would be required going forward, for all re-work performed on Boeing aircraft components ( B737 series Aft Flaps was my work area ).

Before the date specified for the start of documenting all re-work, I felt compelled to try coaxing senior leadership into abandoning the effort before it got started. I shotgun E-mailed
QA and all Management levels with the same message...at the same time.

I pointed out that the time required for workers to input the data and also to call QA for making special inspections... was simply unsupportable and unsustainable. From memory, and Aft Flap Aft Spar was assembled using a combo of 123 total rivits and " Hylock " fasteners. I asked the question of how each fastener requiring re-work could readily be identified ?
For example... how would you inform QA about the location of each & every fastener in question...when they wanted each one to have its own input documentation and inspection?
Example: " 57th rivit in from the Rt outboard edge " ?
Example: " 8th Hylock on the upper row of fasteners joining the spar web to the spar cap " ? The whole shebang would have to have been digitized, and also provide pre-formed wording and / or codes to speed use. Even then...workers would be at the SMS computer for inordinate amounts of time.

I also pointed out that QA inspection levels were already demanding, and new ( increased ) re-work documentation / inspection regimine would have over-obligated QA inspectors; with the result being greatly delayed inspections. To my suprise, QA retracted the message; and things settled back down ! My first level supervisor thanked me for getting management to change their mind.

My second-level supervisor ( who was only seen on the production floor maybe once a year ), made it a point to come down to the floor.... specically to see me. He then proceeded to try and ream me out for the amount of time I must have expended...formulating and obviously very persuasive & successful E-mail message. The day's production needs... making the " rate "... were the paramount concerns, he sternly assured . Figuring he would not take the time to personally find out how much time I had actually spent on the message,
I told him an amount about 1/3rd the actual time I had spent.
And, off he went.

My point: At Spirit, Tulsa.... You could re-work errors you found and/ or made yourself all you needed to.... right up until the task or item was " bought " by QA. Re-work after the item was sold to QA , required prior authorizarion from Engineering and QA.....before such re-work could be legally undertaken ( per QA regs ). Some re-work requires specific Engineering review and aurhorization. For example: when a worker found a " non-shear " rivit ( excess gap rivit head-to-structure), they were allowed to re-drill the hole, and install a .001" or .002"
" oversize " as needed; to straighten out the crooked hole.
However....if the hole was still crooked after trying the .002"
" over " rivit.... they had to obtain Engineering aurhorization to try the last allowable fix.....use of a .003" oversize rivit.

One can see how such items might have contributed to " travel work " in the past, and even so today.


With regards,
357Mag

 

[WatcherZero]

Speaking as a (Pharmaceutical) QA its not so much recording which particular bolt required rework every time, its as much to know that a section has required rework and then we would be looking at our records for the frequency and quantity of reworking required on that section. If we saw a pattern that it was happening a lot (or on certain shifts, or from a particular suppliers materials) we would look at improving the process to be correct first time either through improved practise (better tools/training) or simplification of the task. My suspicion is the reason they wanted supervision of the final allowable fix was because if it failed the part had to be scrapped which was a financial loss for the company, but also the risk workers might not report it failing as they didn't want to be responsible for the loss and so it was passing on and potentially missed during inspection.

Its all about managing risk, one or two minor defects isn't a issue but when there is a large number the cumulative risk of failure makes the part too risky to be used, particularly when lives are on the line if it failed. When these become significant you have to write a justification on why the product is still safe to release to market.

To give an example from my work you might have say in the procedure a maximum number of hours product can wait after initial mixing between moving between along production into a final mixing and then compression, if it broke it for a couple of hours due to a production machine breaking down and requiring repair then you wrote a justification for why that was a minor risk. If the machine was down all week you would have to either write a justification for it being done on a sister production line mixer instead (same equipment model, end result already validated) but you couldn't put it on another machine of a different design that wasn't validated as that would require product stability trials and extensive sampling over a year to validate that it had been properly mixed which would be costly, so you would write that the time interval was now too large and the batch should be binned rather than waiting into the following week. Similarly if a batch was compressed and the pills were too hard or too soft (which affects the speed of digestion in the stomach and dose absorption by the body) which was usually due to varying moisture content from atmospheric conditions you would assess them against the product license (manufacturing tolerances are tighter than the limits in the product license, so if they are still within the license tolerance you could still release them) or if not you could decide whether you want to crush them down and recompress them as a 5 or 10% addition to another different batch to balance out the moisture.

Being a QA for a life sensitive product is a tough balancing act between the interests of the shareholders and the safety of the public, but when you do it right you can not only increase the consumer protection but also find ways of reducing the rejection rate, finding alternative cheaper but same or better quality suppliers and also reducing production material wastage from new processes/altered inputs and saving the company money.