General > General Technical Chat

What did we learn from the "open source ventilator" mess.......

<< < (10/15) > >>

Berni:

--- Quote from: EEVblog on July 08, 2020, 03:07:30 am ---Or perhaps, just perhaps, they needed some good publicity:
https://www.nytimes.com/2020/03/29/business/coronavirus-us-ventilator-shortage.html

--- End quote ---

So the medical equipment vendors are also all greedy unethical capitalists like the pharmaceutical industry.

Then again this is how most big companies do business. That big board of directors does whatever it takes to please the shareholders and investors. They don't care who they piss off as long as the numbers are going up and up.

CatalinaWOW:

--- Quote from: EEVblog on July 08, 2020, 02:19:44 am ---
--- Quote from: james_s on July 07, 2020, 05:24:04 pm ---The Medtronic thing was obviously targeted at companies capable of producing devices of that nature already. If you are already manufacturing equipment and there is a dire enough need, having that design available greatly reduces the amount of reinventing the wheel that needs to be done. Also it was fascinating to be able to look at the design and see just how these things work.

--- End quote ---

I don't understand why Medtronic couldn't just ramp up existing production and pump out all the ventilators needed?

--- End quote ---

This question cuts to the heart of one of the issues.  Medtronic has a design.  And a supply chain that is tooled up and certified at a production rate.  Ramping up production requires either shortening process steps (No you don't have to do a 10 hour burn in on part Z before test.....) or increasing parallelism (adding new suppliers, etc.).  Both choices involve going outside the approved design box, with unknown risks.  Adding parallel lines doesn't churn stuff out faster for a while, it depends on how long it takes to buy and install machinery and the like.  Often all of these steps require certification from some agency which has its own staffing and scheduling problems.

There was a public evaluation of production increases at one vendor of PPE.  His normal rate involved two shifts of production on his line, the third shift being used for repairs and preventative maintenance on the line.  He increased production in the short term by 50% by running production all three shifts.  But was aware that maintenance was going to cause issues sooner or later.  He had his older production line in storage and was bringing it out of mothballs, but there was setup time (measured in weeks as I recall) and it wasn't going to double capacity because the old line had been designed for a lower rate.  He also had to find and train operators for the line as his workforce overtime was already saturated by running three shifts.  Going up beyond that on production was going to require buying all new machinery, hiring and training more people, and was going to require developing suppliers capacity also.  The machinery was all custom stuff, so it wasn't just a matter of opening the appropriate catalog and buying it, it was buying pieces, arranging for fabrication of other pieces, then assembling, tuning and testing the result.

Over multiple months increasing production is realistic.  Doing it in days or weeks is limited.

EEVblog:

--- Quote from: CatalinaWOW on July 08, 2020, 05:38:14 am ---
--- Quote from: EEVblog on July 08, 2020, 02:19:44 am ---
--- Quote from: james_s on July 07, 2020, 05:24:04 pm ---The Medtronic thing was obviously targeted at companies capable of producing devices of that nature already. If you are already manufacturing equipment and there is a dire enough need, having that design available greatly reduces the amount of reinventing the wheel that needs to be done. Also it was fascinating to be able to look at the design and see just how these things work.

--- End quote ---

I don't understand why Medtronic couldn't just ramp up existing production and pump out all the ventilators needed?

--- End quote ---

This question cuts to the heart of one of the issues.  Medtronic has a design.  And a supply chain that is tooled up and certified at a production rate.  Ramping up production requires either shortening process steps (No you don't have to do a 10 hour burn in on part Z before test.....) or increasing parallelism (adding new suppliers, etc.).  Both choices involve going outside the approved design box, with unknown risks.  Adding parallel lines doesn't churn stuff out faster for a while, it depends on how long it takes to buy and install machinery and the like.  Often all of these steps require certification from some agency which has its own staffing and scheduling problems.

There was a public evaluation of production increases at one vendor of PPE.  His normal rate involved two shifts of production on his line, the third shift being used for repairs and preventative maintenance on the line.  He increased production in the short term by 50% by running production all three shifts.  But was aware that maintenance was going to cause issues sooner or later.  He had his older production line in storage and was bringing it out of mothballs, but there was setup time (measured in weeks as I recall) and it wasn't going to double capacity because the old line had been designed for a lower rate.  He also had to find and train operators for the line as his workforce overtime was already saturated by running three shifts.  Going up beyond that on production was going to require buying all new machinery, hiring and training more people, and was going to require developing suppliers capacity also.  The machinery was all custom stuff, so it wasn't just a matter of opening the appropriate catalog and buying it, it was buying pieces, arranging for fabrication of other pieces, then assembling, tuning and testing the result.

Over multiple months increasing production is realistic.  Doing it in days or weeks is limited.

--- End quote ---

Yes, but it's vastly easier and quicker than having other people/companies take their existing design docs and do it all from scratch using new suppliers and processes and personnel etc.
If the government wanted to they could have requested (and paid for) companies like Medtronics to ramp up their output.
Open sourcing their design docs under a very restrictive license was little more than a dog and pony show, and they know it.

Berni:
The difficulty of ramping up production can vary a lot.

Yes a heavily automated high volume production line (such as one that spits out multiple facial masks per second) has no room for speed up. All you can do is make sure it keeps running 24/7 while setting up a second parallel one takes a long time.

But production lines for complex low volume products are not like that (such as a line making high end oscilloscopes or heck medical ventilators). Those are not the same perfectly timed clockwork operations. They might normally make production runs of 100s to 1000s of units where they order the plastic cases, boards, PSUs etc... Once the stuff arrives they get a bunch of people together to assemble it, test it and put it into boxes that go in the warehouse to be sold. Once the warehouse starts running low they push the button for another production run. The same people that put together those oscilloscopes this week might be putting together signal generators the next week.

So on these low volume manual production lines you can ramp up production a lot. The plastic injection moldering supplier just stamps out more ventilator plastic housings rather than washing machine control housings that costumer canceled the order for because there factory is shut down. The people assembling them are not a terribly high skilled force and in a large company you can find workers to move into an extra production shift, you can reduce the production of other products to free up more time on the shared production floor.. etc The only issue might be any specialized test rigs in quality control, they might only have one test rig that is not designed to handle such a volume and a second one would have to be custom built from scratch. But perhaps you can run the test rig now 24/7 while before it might have been run on a single normal 8 hour shift.

tom66:
At the early stages of the disease there was strong evidence to suggest a 10 million+ worldwide death toll; ultimately it looks like it will peter out below 1 million, depending on how the USA and Brazil behave over the next 3 months. 

If the death toll continued to increase, an open source ventilator designed to a minimum, fair standard (certainly not one of those 'bag-squeezing' machines) may have been worthwhile, IF it could be spun up quickly.  But, I think ultimately, it would have been better to take an existing design and mass produce it.   The secondary issue that no one seemed to mention is that nurses, doctors, etc are all trained on a set of specific machines that they'd find in their hospital.  They may need to be retrained to use new equipment, which could take days. Existing equipment is also pre-certified and likely to be safe by design, or at least its limitations are understood.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod