3 ms·

Having done IT work, and helping debug PLCs, for two food startups let me tell you it was a big eye opener. I had this impression that in factories you buy equi

Having done IT work, and helping debug PLCs, for two food startups let me tell you it was a big eye opener. I had this impression that in factories you buy equipment, hook up conveyors and away you go. In reality the equipment is just a starting point. Getting it to produce the end product needs a LOT of trial and error. The smallest adjustments in timing, parameters or recipes can result in a completely different looking, tasting or mouth-feeling product. There is a loop of adjusting the equipment to fit the recipe, then adjusting the recipe to fit the equipment. Then repeat. Lower-end manufacturers also lie about their product specs, lie about the availability of spare parts and MTBF. Saving money ends up costing a lot more, just like it did to Ample Hills. Maybe if you have a department of chemists who are working closely with high-end equipment manufacturers you can better predict the outcome with less trial and error ... but I doubt it. Look at Tesla and the manufacturing issues they are having. The best engineers that money could buy spent a lot of time trying to speed up the production line only to find the hard way that shaving two minutes off time on the line resulted in the paint not having long enough to dry. Food production is a lot harder than software. If you ever want to create a large scale food manufacturer, hire people who have a lot of experience doing it.

macNchz 6y ago
I gained a bit of appreciation for the complexity of this dialing-in process when I was in high school...I was staying in a hotel mostly occupied by my extended family for a wedding, and across the street there was a brand new Krispy Kreme store about to open. During the fine tuning stage of the newly installed machinery they were giving away unlimited free donuts, so my cousins and I formed somewhat of a bucket brigade funneling them across the street whilst piping hot. I ate enough donuts over the course of the weekend that I was able to observe the batches getting more consistent...less variation in color/texture/glaze. I thought it must be interesting to be the person traveling to each new location and making the call as to when everything was exactly perfect.
toyg 6y ago
My ex-wife worked for a massive beer company for a few years, on the back of a degree in advanced chemistry - she was effectively headhunted two years before completing her course. The reality she conveyed was that food manufacturing is really a chemical-reaction business: they have labs full of researchers and quality-controlling chemists, and massive hyper-specialized equipment to manage every aspect of the process, typically unique to this or that plant. The smallest mistake can result in millions of $ of losses, not just from wasted product but from the long effort required to empty and recalibrate a setup. And as you said, machine vendors are as unreliable as, well, any vendor; so you buy these massive things, set them up, and maybe two months down the line you have to call in super-expensive consultants to figure out why a certain sensor is lying to you and spoiling the product.
vmh1928 6y ago
Wasn't there a thread on here a few days ago about some process for making vaccines with this almost exact discussion "just because you buy the <some kind of simple equipment> doesn't mean you can get the whole end-to-end process to run."

macNchz 6y ago

I gained a bit of appreciation for the complexity of this dialing-in process when I was in high school...I was staying in a hotel mostly occupied by my extended family for a wedding, and across the street there was a brand new Krispy Kreme store about to open. During the fine tuning stage of the newly installed machinery they were giving away unlimited free donuts, so my cousins and I formed somewhat of a bucket brigade funneling them across the street whilst piping hot. I ate enough donuts over the course of the weekend that I was able to observe the batches getting more consistent...less variation in color/texture/glaze. I thought it must be interesting to be the person traveling to each new location and making the call as to when everything was exactly perfect.

toyg 6y ago

My ex-wife worked for a massive beer company for a few years, on the back of a degree in advanced chemistry - she was effectively headhunted two years before completing her course. The reality she conveyed was that food manufacturing is really a chemical-reaction business: they have labs full of researchers and quality-controlling chemists, and massive hyper-specialized equipment to manage every aspect of the process, typically unique to this or that plant. The smallest mistake can result in millions of $ of losses, not just from wasted product but from the long effort required to empty and recalibrate a setup. And as you said, machine vendors are as unreliable as, well, any vendor; so you buy these massive things, set them up, and maybe two months down the line you have to call in super-expensive consultants to figure out why a certain sensor is lying to you and spoiling the product.

vmh1928 6y ago

Wasn't there a thread on here a few days ago about some process for making vaccines with this almost exact discussion "just because you buy the <some kind of simple equipment> doesn't mean you can get the whole end-to-end process to run."