Learning Lean

How I used "common sense" to manufacture widgets for TI Automotive.
#Toyota #TIAutomotive


Facebook Share Icon LinkedIn Share Icon Twitter Share Icon Share by EMail icon Print Icon

I will confess that spending an entire day in a seminar is not my favorite way to pass the time. Yet, I spend a fair amount of time trapped in seminars, waiting for someone to say something that's more than marginally interesting. This is pretty much what I thought I'd be doing when I was invited to spend the day at a lean manufacturing seminar that TI Group Automotive Systems (the former Bundy Corp. and Walbro Corp.) was holding for its employees. So I showed up early in the morning, hoping that I could get what I need and sneak out. I even skipped breakfast to put myself in a "lean" mood. In retrospect, I was wrong about two things: (1) This wasn't another boring seminar. (2) You should eat breakfast if you're going to spend the day working in a simulated factory.

In A What?

Lori Casiano, plant manager at the Bundy Hillsdale, MI, facility, is also TI Automotive's lead trainer for its Common Sense Manufacturing (CSM) program. That's right: it's not called "Lean Manufacturing," and there's a good reason for this. As Jim Davis, president of TI Group Automotive Systems explained to me, the word "lean" tends to make people think too much and buy too many computers. He prefers to keep things simple, and that's where Casiano steps in. Under her guidance, we (approximately 35 TI Automotive employees from different departments and disciplines, and I) set up a "plant" in a conference room and ran through three production shifts. Every possible position in the production scenario was represented, from the customer and the plant manager to the production workers and the truck driver.

In the first shift, we operated in a "traditional" push system: we built widgets in batches to maintain inventories. Each workstation performed one procedure on a widget, then material handlers took it to another workstation. In the second shift, we were allowed to "improve" our production process by whatever means we wished, but we still had to build in batches and push product. In the third shift, we completely overhauled the plant to produce the widgets using a CSM pull system.


For the cynical know-it-alls in our group (I won't name names, but if you want a hint, look at the third name on our masthead), the big question before we started the simulation was "Why not just read about this stuff in a magazine to learn it? That's how I learned about lean." The answer, I learned, had a lot to do with doing, not just knowing. To wit, Casiano didn't go into lengthy explanations of the vocabulary before we began the simulation. Once the simulation started, all the vocabulary and textbook understanding in the world didn't matter as much as being able to apply common sense to our specific situation did. Furthermore, the vocabulary is much easier to master once you've completed the simulation. Ergo, the simulation is more effective than a lecture. As Davis says, "This is not a linear exercise." It must be experienced. (Not to mention that for those few souls like myself who hate lectures, a simulation is a lot more fun.)

So What Happened?

As the old saying goes, "There's more than one way to skin a cat." Casiano explains that no two simulations play out the same, but every one that she administers has certain similarities:

  • The first shift is, at best, chaotic. It results in poor productivity, an unhappy customer, and the plant manager is usually hated by the employees and/or fired.
  • The second shift is less chaotic. Workers always figure out a way to produce more widgets and better satisfy the customer, but inventory goes sky high. The plant is more productive, but loses even more money than the first round due to excess inventory.
  • The third shift proves the point. By implementing a just-in-time system with kanbans (the starting point for CSM; this is based on the universally accepted model of lean manufacturing: the Toyota Production System), productivity increases, inventory is eliminated, the customer is delighted, and the plant makes money.

The differences from simulation to simulation are in the implementation. Different groups of people choose to organize their simulated plants differently. But every group finds a way to implement CSM effectively in the third shift. That's a key point: there is no single "right" way to become lean. This is because CSM/lean focuses on the cause, not the effect. If the process is improved, so is the outcome; there are always many ways to improve the process.

This principle, however, can be a major stumbling block. Many manufacturing people have a hard time trusting in a lean philosophy enough to forget about the numbers and focus on improvement. "Problems are good!" Davis says, because they give you opportunities to fix what's wrong with your organization. The alternative to fixing problems is usually to "work harder" at the problem in the same way as you were before. In manufacturing, this usually means adding production capacity (either machinery or workers) which is used to build more inventory. This is what we did in the second shift of our simulation. It appeared to work great, except that the cost of all that inventory prevented us from making any profit. On the surface, things looked good (at least they looked better than before!), but the inventory was hiding waste.

Eliminating waste seems like an obvious enough goal, but waste can come in so many different forms that its elimination is a never-ending process. This is kaizen. Inventory in itself is waste, but so are rework, down-time, setup time, poor communication—even the failure to take advantage of the skills of workers can be considered waste. Perhaps the biggest lesson in eliminating waste that we learned from the CSM simulation is that much of material handling is waste. Since handling raw materials and work-in-process doesn't add any value to the product, it's not helping create profit. Therefore, we streamlined the flow of widgets through our plant by measuring takt times and rearranging our production line so that we were handling in-process widgets a lot less. Our other great accomplishment in eliminating waste was using jidoka. We designed fixtures at each workstation to assemble and check the parts for quality as we built them. Therefore, none of the parts that we made had to be reworked (and we satisfied our QS9000 requirements).

"So What?"

What do you say to the skeptic? It's easy to reject the entire notion of lean or CSM based on the grounds that it's irrelevant because, "We don't do that here," or "We don't need to do that because we're making money anyway." To this Davis and Casiano reply with one word: "Customer." The CSM simulation makes "The Customer Is King" a very easy philosophy to understand. The pull system forces each step of the production process to regard the operation immediately following it as its "customer." Responsibility for customer satisfaction is thus spread throughout the manufacturing process. Among production workers this leads to a higher sense of ownership in the work and a higher level of satisfaction in the job. These are good things; they lead to better productivity and further improvements in the production process.

When you follow the process, from customer to customer to The Customer, it becomes fairly obvious why lean works so well in achieving customer satisfaction and profits. To understand this relationship, consider the first rule of the auto industry. My high level sources explain it quite simply: "Make money." This, by definition, gives The Customer a cocky swagger that is to be respected, as the customer is the one paying the bills.

Get It

While the simulation was by no means a comprehensive education in lean (that's why TI Automotive follows it up with a second in-plant workshop), it made the pieces of the lean manufacturing puzzle fit much more tightly together in my mind. While Davis and Casiano didn't pen the concluding statement below, I think they'd still have given me that diploma if I had written it on my post-simulation quiz.

All companies get lean in one of two ways: Either they get lean and make a lot of money or they get lean because they don't make a lot of money.

TI Automotive's "Common Sense" Glossary

Jidoka: A system to ensure quality, whereby machines are built so they can only produce quality parts.

Just in time: Parts are made per the customer's order, one at a time, not in batches.

Kaizen: Continuous improvement, meaning eliminating waste and non-value-added work. "There is always a simpler way."

Kanban: A visual pull system. For example, a rack might contain four variations of a part. When a part is pulled from the rack for a customer, the operator builds another identical part to fill the empty spot in the rack.

Push system: The way McDonald's has always operated (prior to their "Just for You" program) is a traditional "push" system. They make lots of burgers and stock them under heat lamps. When you order, they give you a burger from stock.

Pull system: Wendy's is a "pull" system. They don't make burgers until they're ordered. When you order, they make your burger from the raw materials.

Takt time: The allowable time to produce one product at the rate the customer needs. Each workstation in a production line should perform its operation within this time.