Industrial improvement systems — strategies for making a business more profitable, more efficient, or better operated — rise and fall in popularity over time.
I work in this space, and I often see people try to group parts into “part families” and process them in some sort of bunched manner. The hard part with the whole premise is that tries to optimize high mix, low volume product but that product is too unstable in design or too low volume to see returns on process optimization.
If I run a machine shop and part A comes from customer A and part B comes from customer B, and the parts have a lot of similar operations that need to be completed, grouping them physically as they travel the shop floor requires 1) the upfront planning, 2) clear and similar production runway from customers A and B, and 3) that there’s something to be gained vs just running a bigger batch of each part. Also from a quality perspective, it would be a nightmare to try to actually mix product in a chemical bath or any specific step of a manufacturing process that you would be able to.
Now you could argue that it’s less about batching and more about factory layout, but then I think you get away from the core premise and you’re just back to making simple conclusions like “I should put my deburr station in between my milling and inspection stations.”
The place I think family building does provide a return is if customer A realizes that 10 of their parts are all pretty similar and they give all that work to one machine shop. That machine shop might not literally bunch them along the production process, but their expertise in developing the recipe for one part will extend to the rest and they’ll be able to offer better prices seeing the larger runway and work volume coming their way.
Work Simplification was pushed by the federal government during WWII and lasted a decade or two longer in government than in industry. The same is broadly true of group technology, which was called "layout studies" in the US jargon. It wasn't nearly as developed as in the USSR but nonetheless was definitely important.
The military applied it to arsenals and shipyards, but even the civilian agencies applied it to improving the flow of paperwork through an office. Here's a random brief illustration from a military manual (pg 27–30):
I've written quite a bit about where work simplification came from and what happened to it. Layout studies were part of the same story. In all cases, of course, these approaches drew upon the quite fleshed out approaches to manufacturing that factories had developed. Perhaps of interest:
Maybe the collapse of the Soviet Union and the very public exposure of soviet economic woes ruined the prestige of Group Technology. Business strategies are strongly subjected to cycles of hype and buzzword abuse so when the Soviet Union was rapidly industrializing it was a model worthy of emulating but after the fall of the Eastern Block the Toyota Way became more cool than the Chernobyl Way.
This is an interesting point, although it's my sense that Group Technology lost favor in the Soviet Union as early as the 1970s, and books on it often tend to downplay or omit entirely its Soviet history.
At my facility we essentially do this tho I don’t think it was based on thei theory, it’s far more efficient to laser cut an entire sheet down to a skeleton than to cut parts as they enter the system so any given day new parts are added to a pile based on material and gauge then cut at once, similarly after deburing which is universal they get sorted by process rather than final product, alone machine may have its part split into the press pile shear tapping welding etc, the sheet metal part of This came from engineering (they used to cut first come first out) but all the rest was just the most efficient way the shop found to organize themselves.
This reduces handling, tooling, and cycle times, optimizes equipment utilization, and reduces plant costs and size, achieving almost the benefits of mass production
Student-centered educators across the globe should rejoice with this post!!! We know that memorable learning IS a form of Group Technology. Thanks for the injection of rigor and curiosity sparking.
thank you for sharing, I think some of your points are valid! Guess neighbour competitiveness in manufacturing drow it out of the western markets, similar to the cover-my-ass specifications overhead? In USSR and now RU, and CN, that cover-my-ass is not a real concern but may be becoming it (anyway there could be a marketplace need, for companies to expose their capabilities instead of the products, much like the lean set based decisions? and maybe AI could help cover-the-ass?)
I think there forever will continue to be a need for grouping, both as a pull when manufactures grows but still only makes less than 1000 identical items, and as push since skills are automatically grouped in the workforce dues to cognitive load.
We don't want to see 10k+ identical buildings, & in construction you spend most of the time identifying items that can be used for the construction, hence many people have id'ed that this industry has seen least innovation since 1960s.
Before 2000 shipyards side-loaded cruise ship prefab cabins, & we are partially seeing a return to western countries, with prefab'ed hulls delivered from the east.
I hadn't heard of Group Technology previously but my intuition is that its value compared to Lean falls as setup time decreases as proportion of WIP time.
Isnt modular design and engineering the surest way in "engineered to order" products? In the 21st century , most machining is high programmable and reconfigurable so group technology doesnt really provide the same set of benefits in manufacturing
Fascinating reminder that improvement systems are as much about branding and narrative as they are about actual results.
Group Technology delivered many of the same efficiencies we now credit to Lean, but its focus on complex coding and documentation made it less marketable
Very interesting post. Grouping sounds like a lot of sense for the world outside the big assembly lines. You mentioned the trend of codifying as much as implementing.. such a human tendency. So much simpler than taking on working out what the best outcome is, and optimizing for that.
I work in this space, and I often see people try to group parts into “part families” and process them in some sort of bunched manner. The hard part with the whole premise is that tries to optimize high mix, low volume product but that product is too unstable in design or too low volume to see returns on process optimization.
If I run a machine shop and part A comes from customer A and part B comes from customer B, and the parts have a lot of similar operations that need to be completed, grouping them physically as they travel the shop floor requires 1) the upfront planning, 2) clear and similar production runway from customers A and B, and 3) that there’s something to be gained vs just running a bigger batch of each part. Also from a quality perspective, it would be a nightmare to try to actually mix product in a chemical bath or any specific step of a manufacturing process that you would be able to.
Now you could argue that it’s less about batching and more about factory layout, but then I think you get away from the core premise and you’re just back to making simple conclusions like “I should put my deburr station in between my milling and inspection stations.”
The place I think family building does provide a return is if customer A realizes that 10 of their parts are all pretty similar and they give all that work to one machine shop. That machine shop might not literally bunch them along the production process, but their expertise in developing the recipe for one part will extend to the rest and they’ll be able to offer better prices seeing the larger runway and work volume coming their way.
Work Simplification was pushed by the federal government during WWII and lasted a decade or two longer in government than in industry. The same is broadly true of group technology, which was called "layout studies" in the US jargon. It wasn't nearly as developed as in the USSR but nonetheless was definitely important.
The military applied it to arsenals and shipyards, but even the civilian agencies applied it to improving the flow of paperwork through an office. Here's a random brief illustration from a military manual (pg 27–30):
https://www.google.com/books/edition/Work_Simplification/NxQ02QmmywkC?hl=en&gbpv=1&pg=PA27&printsec=frontcover
I've written quite a bit about where work simplification came from and what happened to it. Layout studies were part of the same story. In all cases, of course, these approaches drew upon the quite fleshed out approaches to manufacturing that factories had developed. Perhaps of interest:
https://www.statecapacitance.pub/p/how-the-feds-invented-lean-manufacturing
Maybe the collapse of the Soviet Union and the very public exposure of soviet economic woes ruined the prestige of Group Technology. Business strategies are strongly subjected to cycles of hype and buzzword abuse so when the Soviet Union was rapidly industrializing it was a model worthy of emulating but after the fall of the Eastern Block the Toyota Way became more cool than the Chernobyl Way.
This is an interesting point, although it's my sense that Group Technology lost favor in the Soviet Union as early as the 1970s, and books on it often tend to downplay or omit entirely its Soviet history.
At my facility we essentially do this tho I don’t think it was based on thei theory, it’s far more efficient to laser cut an entire sheet down to a skeleton than to cut parts as they enter the system so any given day new parts are added to a pile based on material and gauge then cut at once, similarly after deburing which is universal they get sorted by process rather than final product, alone machine may have its part split into the press pile shear tapping welding etc, the sheet metal part of This came from engineering (they used to cut first come first out) but all the rest was just the most efficient way the shop found to organize themselves.
I sense an analogy to the Haier Way:
https://www.corporate-rebels.com/blog/rendanheyi-forum
In that it's breaking manufacturing into isolated components that need to contain their own customer focus and quality control.
This reduces handling, tooling, and cycle times, optimizes equipment utilization, and reduces plant costs and size, achieving almost the benefits of mass production
Student-centered educators across the globe should rejoice with this post!!! We know that memorable learning IS a form of Group Technology. Thanks for the injection of rigor and curiosity sparking.
thank you for sharing, I think some of your points are valid! Guess neighbour competitiveness in manufacturing drow it out of the western markets, similar to the cover-my-ass specifications overhead? In USSR and now RU, and CN, that cover-my-ass is not a real concern but may be becoming it (anyway there could be a marketplace need, for companies to expose their capabilities instead of the products, much like the lean set based decisions? and maybe AI could help cover-the-ass?)
I think there forever will continue to be a need for grouping, both as a pull when manufactures grows but still only makes less than 1000 identical items, and as push since skills are automatically grouped in the workforce dues to cognitive load.
We don't want to see 10k+ identical buildings, & in construction you spend most of the time identifying items that can be used for the construction, hence many people have id'ed that this industry has seen least innovation since 1960s.
Before 2000 shipyards side-loaded cruise ship prefab cabins, & we are partially seeing a return to western countries, with prefab'ed hulls delivered from the east.
I hadn't heard of Group Technology previously but my intuition is that its value compared to Lean falls as setup time decreases as proportion of WIP time.
As a bit of a process nerd, I enjoyed this
Isnt modular design and engineering the surest way in "engineered to order" products? In the 21st century , most machining is high programmable and reconfigurable so group technology doesnt really provide the same set of benefits in manufacturing
Fascinating reminder that improvement systems are as much about branding and narrative as they are about actual results.
Group Technology delivered many of the same efficiencies we now credit to Lean, but its focus on complex coding and documentation made it less marketable
Very interesting post. Grouping sounds like a lot of sense for the world outside the big assembly lines. You mentioned the trend of codifying as much as implementing.. such a human tendency. So much simpler than taking on working out what the best outcome is, and optimizing for that.
Do you have any good books on this you can recommend?