For years I’ve been reading the Algorithmics Anonymous forum, eagerly anticipating the discovery of a new form of lean that applies only to "high mix low volume" manufacturing operations. I continue to wait, and it appears that many others are starting to get a little impatient as well…. about every week there is someone asking, or re-asking, "could this [insert topic du jour here] be our focus?"
Get used to it. The group vascillates between trying to find an almighty algorithm solution to the unnecessarily complex scheduling problems of high mix operations and simply bashing "Toyota Lean" as not being applicable to such operations. But over the years there has not been a single "a-ha!", and there probably never will be… because they are trying to identify a subset derivation of something that is already very simple. Which is why out of many hundreds of members, there are about five or ten that actually post… and endlessly debate the same inconsequential points.
AlgoAnon tries to differentiate "assembly operations" and "job shops", but by doing so they miss one of the fundamental concepts of lean (TPS-lean, whatever): achieving one piece flow. As Bill pointed out a few weeks ago, as you approximate one piece flow, even when building cars, the "volume" side of "high mix low volume" or "high volume low mix" goes out the window, thereby rendering the argument irrelevant. This does not turn them into identical operations… they will always be optimized for certain types of products and operations. But the continuous improvement aspects become analogous.
Algorithms can be fun to play with. Trying to model manufacturing operations is an interesting challenge. But try as you might, you can never perfectly predict the unpredictable or account for the unaccountable. They can help you optimize, but they can be dangerous in a control situation. The simplest manufacturing operation can be thrown out of kilter by the simplest of unexpected delays… let alone a highly complex HMLV operation reliant on widely disparate processes, deliveries, and suppliers. The collaborative self-management and self-correcting aspects of simple intuitive visual displays can be far more powerful and agile in reacting to such circumstances than the most complex software "solutions." Probably without knowing, the group argued that same point:
Toyota has mastered the thinking underlying this profession [IE, lean] while we in the U.S. got all taken up with math and MBA managers. Go back to basics.
Yep, sounds like how we came up with all them there algorithms.
But what I find most dangerous is the idea that lean and Toyota-lean can’t work in the job shop. How many job shops are buying that line, not bothering to improve waiting on the holy grail of the perfect ultimate solution, and are being forced out of business by domestic and overseas competitors who have their heads on straight? Some of the comments from just last week alone, which I’ll keep anonymous in the spirit of AlgoAnon, are downright scary:
TPS only offers a vision, a tool here and a tool there, it offers nothing comprehensive.
The Toyota-inspired lean is the wrong route! An outside guru can go in and do his things, like VSM and SMED and 5S. But with what long term effect?
Adaptability, flexibility, forecasting in the case of uncertainty… what has Toyota produced in this arena?
Value stream maps and lean assessments that are catering to high-volume assembly, and the focus that assembly cells provide, are not applicable to the typical jobshop.
Continuous improvement as done at Toyota is not what a jobshop should adopt. It would set the jobshop up for a prolonged transformation, often totally wrong.
Do I believe that Toyota’s version of lean is the only true lean, or even that lean is the only way to achieve excellence? Of course not. Toyota itself is always improving, and there will always be new and improved ways of doing things. But do I believe that Toyota’s version of lean can create tremendous value for all forms of manufacturers, and even non-manufacturers? Yes.
Perhaps the best way to address this nonsense would be to take on the challenge that the group leader posted:
Ok, so you have this small shop owner that wants to improve, and then improve continuously. Now what? Value stream mapping? 5S? Download the lean assessment tool that Eaton and Lockheed Martin use and use that verbatim?
Yes, in a methodical and planned manner.
First, just start learning. Get the knowledge, then teach it to your folks. Learn about waste and how to recognize it from the perception of the customer. Waste is waste… nothing job shop specific here. Send people to AME conferences and hands-on workshops at real manufacturers. You’ll probably find that the desire and enthusiasm for change has been there all along… they just needed to be led.
Second, examine your organization and operation. Yes, assessments can help focus that examination and provide a baseline and benchmarks. I personally like Brian Maskell’s Journey to Agile Manufacturing assessment tool as it clearly describes the environment at each level, which will often further open the eyes of people who take it. Again, nothing job shop unique.
Third, create a plan. Look at the result of the assessment, think about what you’ve learned, review what you’ve seen at other plants, consider hoshin kanri. Involve everyone in the plan. Create goals for the next year and five years. Determine what learning and perhaps even outside resources are required. Share the plan, implement it, and review and revise it periodically.
Fourth, yes do implement appropriate tools. VSM does not need to be applied specifically to a specific manufacturing line (or job shop process area…). Feel out the concept by mapping how you take and process orders… a process common to all manufacturers, and most non-manufacturers for that matter. If you are like 90%+ of companies you will immediately see how unnecessarily complex the process is, and you will typically be able to streamline it by about 30% on the first pass. Do it again somewhere else. Implement 5S… clear out unused equipment, tools, put things in a clear place. This helps everyone… even outside of manufacturing. Implement visual factory… post metrics, information, and status… even (egads!) try scheduling on a whiteboard in the middle of the production floor instead of relying on specialized people squinting at computer screens. Suggestion programs, root cause analysis, rewards. Continually focus on reducing waste in all forms… you’ll be surprised with the dark corners where it is found (often literally).
Continuous improvement done right has a habit of becoming infectious. You’ll be able to feed on the energy. You’ll be able to liberate great ideas. Best of all this has nothing to do with whether you’re an "assembler" or a "job shop"… or even a manufacturer at all.
AlgoAnon’s moderator often asks
Does it not bother you that you get what you are learning from somebody else [Toyota]? Do you not wish you were the person teaching others something that you invented?
Nope, I don’t care in the slightest. Who cares about NIH? It does not bother me to learn from others. My only desire is to help others succeed. Do we simply want to copy Toyota? Of course not. That’s why manufacturing executives are paid the big bucks… to lead and manage and find the right solutions. Any street corner consultant can copy.
Sure there will be differences… between "assembly" and "job shop" just like there are between "hospital" and "government" and "manufacturer". But the relentless pursuit of perfection, waste elimination, and employee involvement is the same. And by the time you get to the point where you’ve driven out the 90% of waste that is common to almost any organization and need to focus and optimize for the remaining 10% that is due to your specific niche you will be so hypercompetitive that it will be basically a non-issue.
Just do it. Just start down the path. By trying to convince job shops that traditional lean won’t work, AlgoAnon is sacrificing companies at the altar of the almighty algorithm.
Bill Waddell says
“TPS only offers a vision, a tool here and a tool there, it offers nothing comprehensive.”
The commentor’s ignorance of lean is appalling. It is not that lean offers nothing comprehensive – it is only that the commentor does not have a comprehensive knowledge of lean.
“The Toyota-inspired lean is the wrong route! An outside guru can go in and do his things, like VSM and SMED and 5S. But with what long term effect?”
If your guru only gives you VSM, SMED and 5S, you need a new guru. (See note #1 – I recommend that you and the guru get a comprehensive understanding of lean.)
“Adaptability, flexibility, forecasting in the case of uncertainty… what has Toyota produced in this arena?”
Toyota has written the book on Adptability and Flexibility (ignorance of this fundamental point goes back to #1 – i.e. a failure to have a comprehensive understanding of lean.) Toyota and the Lean community are oblivious to forecasting tools because forecasting is a fool’s errand. All of the science, engineering and supercomputing in the world driven by real time satellite imagery cannot even reliably tell me if it’s going to rain tomorrow. Who is it that thinks they can predict the sales of widgets months off into the future?
Refer again to the ignorance of lean in a comprehensive manner. Lean is based on the premise that forecasting will never be accurate and the key to manufacturing excellence is to be ever more reactive to whatever happens in the market. Better forecasting to enable push manufacturing to succeed is a dream only Don Quixote can love.
“Value stream maps and lean assessments that are catering to high-volume assembly, and the focus that assembly cells provide, are not applicable to the typical jobshop.”
No kidding? I suggest that you not use them if they don’t apply. However, don’t confuse the fact that every factory is better suited to some lean tools than others is hardly a reason to abandon lean. If you think that lean is dependent on the deployment of any particular lean tool, we are back to #1 again.
“Continuous improvement as done at Toyota is not what a jobshop should adopt. It would set the jobshop up for a prolonged transformation, often totally wrong.”
I would suggest that any job shop that decides “continuous improvement as Toyota does it” is “totally wrong” will not be around a whole lot longer, so the point will sooon become rather moot.
Thankfully, you are right, Kevin. The modren day equivalents of buggy whip makers railing against these newfangled automobiles are few in number and, in the end, of harmful consequence only to their employers and the few others who cannot see that their opinions are driven by a tenacious desire to hang on to the past and an inexcusable ignorance of manufacturing.
Micheal Gardner says
I run into this kind of person from time to time and I have come to the conclusion that those who go out of their way to “prove” that “it won’t work here” are doing so out of fear. They are more afraid of changing than of losing their job or business. Fear can be a powerful motivator and can cause otherwise rational people to invent elaborate scenarios justifying their actions (or inactions) that are actually unreasonable.
On the flip side, I have worked with some of our company’s suppliers who are “job shops” and have incorporated much of the lean enterprise philosophy. A motor remanufacturer we work with typically schedules production as “one of these, then five of those, then two of these, then two of these, etc…” They are able to capitalize on their extreme flexibility as a business advantage. If you need a starter for a 1965 Fiat sedan right now, you will get it “right now” as constrained only by FedEx, not their ability to quickly manufacture it. They keep only minimal inventories of raw materials on hand and virtually no finished products. Yes, they have an elaborate computer system with specialized software, but most of it is used for communicating technical specifications and drawings directly to the work cell operators and other information of that nature. And I’m sure the owners/managers of this company have never bothered to read AlgoAnon.
Karen Wilhelm says
It’s been known for ages that job shops typically make many things that are similar. Thus “group technology,” a term that has at least two meanings, joins VSM to allow cells to be formed around a segment of work that can be treated as higher volume, lower mix. The sequence of operations for the parts made in low volume is analyzed and those that have the most similar sequences are grouped together. Part #1 flows through the cell with operations A, B, C, E. Part #2 undergoes operations B, C, D, and so on. The cell is still scheduled, but is more flexible. There will still be one-of-a-kinds, and very low volume jobs that don’t fit into the pattern, but the situation is much simplified. You don’t need an algorithm to do the analysis, just a few sheets of graph paper, a simple spreadsheet, or process planning paper.
JS says
I see someone just posted the entire text of this post on the JSLean forum. That should generate some excitement!
I am one of the silent majority that reads the forum but hasn’t learned a thing except that several people can’t move off of square one. However I do think there are some unique aspects of job shops, but in general I agree that pretty much all of TPS can be applied to job shops in some form, and that huge benefits can be derived from it.
Your last point is the most important: just do it. By the time you get to the point where the differences between job shops and traditional assembly operations play a part in productivity improvements you will already be so competitive that it won’t matter much. Huge productivity and competitiveness gains are possible by just focusing on the basics of lean.
Thanks for continuing to challenge our status quo.
JS
Currently on step 9 of 12 at AlgoAnon.
Jon Staughton says
TPS works perfectly fine in job shops. Anyone that thinks otherwise does not understand TPS or is a crackpot consultant trying to make a buck.
I work at a small shop that makes everything from mechanical to electrical to pneumatic subassemblies all on our one “line” with no computers. We have no idea what we will make from week to week. But we’ve reduced inventory and used TPS to greatly improve flexibility and agility which is now our primary selling point.