Lean Centaurs: The Future of Process Improvement

Lean Centaur

I’ve been reading a fascinating book this 4th of July weekend: The Inevitable: Understanding the 12 Technological Forces That Will Shape Our Future, and trying to relate what I’ve learned to the world of Lean and Process Improvement. I’m still making my way through the book, but I have enough raw material to get started on these speculations. There are enough ideas for quite a few posts, but here’s a beginning.


The movie The Terminator brings to mind the dark side of Artificial Intelligence, computer logic at stark odds with humanity. Skynet, the computerized global defense system, becomes self-aware and embarks on the destruction of the human race. And there are some famous names, including Steven Hawking and Elon Musk, who have voiced concern about this possibility. Check out an Open Letter that they both co-signed emphasizing the need for caution and careful research.

The impression I get from The Inevitable is quite a bit more benign, however. The AI systems in use today and in the near future are not “conscious”, although scientists are working on this possibility. Instead, the AI systems are a form of Advanced Google Search, that can respond to queries of various types, and respond with recommendations and answers that a human being would be hard-pressed to accomplish, certainly quickly. For example: input symptoms and lab results regarding a physical illness, and the AI system can respond instantly with diagnosis options and treatment plans based on the latest medical research. Personally I would want a human doctor between me and the AI, but acting in the role as a quality control manager. Can my doctor keep up with all of the advances in medical research, without using an AI tool? Probably not.

So how might this apply to the world of Lean and Continuous Improvement? Picture this: a Manufacturing Engineer(s) populates a database with process-related information, like products, processes, process times, process flow diagrams, target volumes, sales history. Add to this measurements of process variability, and maybe even a CAD drawing showing the available floor space. Click on a button, and the “Lean AI” returns a proposed layout, along with an optimized sequencing plan, staffing plan, and capital acquisition budget. An AI system does not even need to know how to do all of this, initially. It can learn over time, after being exposed to many examples and iterations. In case you’re thinking that this is a bit far-fetched, look at what Google is doing to develop their own in-house AI chips.

Google’s chips can be slotted into the company’s computers, rapidly augmenting them with faster artificial-intelligence capabilities. It represents a first attempt by Google at designing specialized hardware for its AI workloads. As the field matures, Google “might very well” build more specialized processors for specific AI tasks.
As with a medical diagnosis, I would want an ME or IE standing between an AI design and physical reality, but this scenario does not sound impossible. It’s already happening, on a simpler level, with the use of computer simulation tools.


A Centaur is a mythological creature that typically involves the merging of the body of a horse and the torso of a human being. Peter Griffin from the cartoon Family Guy envisioned himself as a centaur, of a slightly different type. The basic idea is that of combing two very different creatures into one.

A “Lean Centaur” is the merging of a Lean Specialist with Lean AI tools. A Lean Specialist without AI tools is going to be less efficient and effective in his/her work, especially as it applies to disciplines like Line Design, Material Management, Kaizen. Sure, a Lean Leader needs to have “soft” skills as well, and maybe those skills should be considered as “hard”. But let’s face The Inevitable: the technical aspects of Lean will be automated and enhanced with AI, and this is happening now, and rather quickly. The Lean Centaur is the creature that brings a human touch to the AI world.

If The Inevitable is to be believed, the world of manufacturing in general will be transformed. Not only will existing assembly workers be replaced completely, but also warehouse and material handling functions will increasingly be automated with inexpensive robots. Again, in case you think that this is a far-future fantasy, look at what Foxconn has already done in their Chinese factories. The implications for Lean are far-reaching, and they point to an increased need for good Lean design skills, in which “Kaizen” needs to take place up-front through tools like Simulation Modeling rather than at a later time in the Value Stream’s life. There’s a lot less room to “fix it” after the automated line is in place.


The main difference between a “normal” Lean person and a Lean Centaur is the ability to use advanced tools. My main recommendation is this: get really good at using the tools that are available to you right now, and keep up on advances over time. The most important tool available to you right now, in addition to a high level of expertise with the Office suite, is the use of Simulation Modeling. Yes, there is a learning curve required, but being able to create models is the #1 way to set yourself apart from the crowd. Check out iGrafx or Simul8 for two easier-to-use software options.

The other skill requirement for a Lean Centaur to know and use are the step-by-step Line Design and Material Management Roadmaps. The software tools, at least today, are not intelligent, and rely on the user (you) to apply them correctly.

Leave a Reply