Why Creating Flow Is Difficult

So, life is good and all your problems and pains simply disappear by creating one-piece flow cells. Not by a long shot! In lean thinking, life will get tougher for a while—at least until you learn how to continuously improve your processes. Ohno explained:

In 1947, we arranged machines in parallel lines or in an L shape and tried having one worker operate three or four machines along the processing route. We encountered strong resistance among the production workers, however, even though there was no increase in work or hours. Our craftsmen did not like the new arrangement requiring them to function as multi-skilled operators. They did not like changing from one operator, one machine to a system of one operator, many machines in different processes. Their resistance was understandable. Furthermore, our efforts revealed various problems. As these problems became clearer, they showed me the direction to continue moving in. Although young and eager to push, I decided not to press for quick, drastic changes, but to be patient. (Ohno, 1988)

One of the comforting things about traditional mass production thinking is that if any one process stops producing—because it takes a lot of time to changeover a machine to a new process, or because a person is out sick, or because the equipment breaks down—the other “separate” process operations can keep on working because you have lots of inventory. When you link operations together in a one-piece flow, your entire cell goes down if any one piece of equipment fails. You sink or swim together as a unit. So why not have some inventory to make life a little more comfortable? Because whether it is a pile of material or a virtual pile of information waiting to be processed, inventory hides problems and inefficiencies. Inventory enables the bad habit of not having to confront problems. If you don’t confront your problems, you can’t improve your processes. One-piece flow and continuous improvement (kaizen) go hand in hand! If your competitor challenges itself to adopt the pain and “confusion” of lean thinking, you’ll no longer be comfortable hiding behind your inventory—you’ll be out of business. As Minoura, former President of Toyota Motor Manufacturing, North America, and a disciple of Taiichi Ohno, explained:

When they run one-piece production, they can’t have the quantity that they want so everybody gets frustrated and doesn’t know what to do. But then, within that, they have to find ways to think: what is the way to get the quantity? That is the true essence of TPS and, in that sense, we create confusion so we have to do something different in approaching this problem.

Many companies I have visited make one of two mistakes when implementing flow. The first is that they set up fake flow. The second is that they go backwards from flow as soon as problems occur.

An example of fake flow would be moving equipment close together to create what looks like a one-piece flow cell, then batching product at each stage with no sense of customer takt time. It looks like a cell, but it works like a batch process.

For example, the Will-Burt Company in Orrville, Ohio, makes many products based on steel parts. One of its larger-volume products is a family of telescoping steel masts that are used in vans for radar or for camera crews. Each mast is different, depending on the application, so there is variation from unit to unit built. This company called its mast-making operation a cell and believed it was doing good lean manufacturing. In fact, before I helped lead a lean consulting review of their processes, a production manager warned us that, with the variety of parts they made, we would not have any luck improving the flow.

In a one-week kaizen workshop, the current situation was analyzed and this turned out to be a classic case of fake flow. The work time (value added) it took to build one of these masts was 431 minutes. But the pieces of equipment for making each mast were physically separated, so forklifts were moving big pallets of masts from work station to work station. WIP built up at each station. With the WIP, the total lead time from raw material to finished goods was 37.8 days. Most of this was the storage of tube raw material and the storage of finished goods. If you just looked at the processing time in the plant, it still took almost four days from sawing to final welding to do 431 minutes of work. The travel distance of the mast within the plant was 1792 feet. The solutions were moving the equipment closer together, moving one piece at a time through the system, eliminating the use of the forklift between the operations (a special dolly had to be created to move this large unit at workstation height between two of the operations that could not be placed next to each other), and creating a single shop order for one mast instead of batches of shop orders for a set of masts. The results of these changes were significant improvements in lead time , reduced inventory, and reduced floor space.

One side benefit of the workshop was that the time to set up a shop order was investigated. The batching of shop orders was creating a lot of waste; when that system was eliminated, it reduced this time from 207 minutes to 13 minutes.  Depicts the process flow before and after the one-week kaizen workshop. You can see that the “before” situation was really a case of fake flow. Pieces of equipment were sort of near each other, but there was not really anything like a one-piece flow. And the people working in the plant did not understand flow well enough to see that it was fake flow. The “after” situation is a marked improvement that surprised and delighted everyone in the company. They were shocked that it could be done in one week.

The second mistake when implementing flow is backtracking, which occurs as soon as the company realizes that there may be a cost to creating flow. This extra cost can occur in any of the following situations:

  • There is a breakdown in one piece of equipment, causing the whole cell to stop production.
  • A changeover of another piece of equipment takes longer than expected and delays the whole cell, stopping production.
  • To create flow, you must invest in a process (like heat treating) that is currently being sent outside to a supplier and bring it in-house.

In all these cases, I have seen some companies give up on flow. Their thinking is that flow is a good idea when we do a simulation with toy parts to illustrate the benefits of reducing batch size and creating flow. But it is a bad idea when we actually try it in a real operation and it causes some short-term pain and cost. When a cell is set up, it takes discipline to maintain, which often is beyond the capacity of many manufacturing companies, because they don’t understand the challenges and pain of continuous improvement. In the long run, the challenges and pain and short-term costs almost always produce dramatically better results.

Toyota’s focus in any process is always on creating a true one-piece flow system that is waste free, as embodied in Toyota Way Principle 2. Create continuous process flow to bring problems to the surface. Creating flow means linking together operations that otherwise are disjointed. When operations are linked together, there is more teamwork, rapid feedback on earlier quality problems, control over the process, and direct pressure for people to solve problems and think and grow. Ultimately, within the Toyota Way the main benefit of one-piece flow is that it challenges people to think and improve.

Because of the focus on thinking, TPS is now being referred to by Toyota as the “Thinking Production System.” Toyota is willing to risk shutting down production in order to surface problems and challenge team members to solve them. Inventory hides problems and reduces the urgency to solve them. The Toyota Way is to stop and address each problem as it is exposed.  Build a Culture of Stopping to Fix Problems, to Get Quality Right the First Time (on jidoka) explains this in more detail.

Case Example: Job Summaries in a Navy Ship Repair Facility

An excellent example of one-piece flow in a service operation was implemented at Puget Sound Naval Shipyard in fall 2001. At this shipyard they do not build ships, but repair Navy vessels from submarines to carriers of various kinds. Every repair situation is unique, so there is heavy involvement of engineering in diagnosing the problem and writing work instructions for the repair. The engineering documentation, including work instructions, is compiled in a folder that goes to the shipyard so skilled mechanics can conduct the repair. This leads to the authorization, funding, and paperwork that mechanics use to do the repair job. The work instructions folder had become a bottleneck in the planning process, in many cases, as well as a cost driver.

To improve the process, a one-week kaizen workshop was run. There was a good deal of preparation leading up to the workshop, including preparations for a reorganization and office space assigned to the creation of a cross-functional cell to process work instructions. The workshop focused on mapping the current process and developing a new process. The wastes identified by analyzing the process step by step included rework, redundant systems, different communication vehicles (e.g., spreadsheets), waiting for forms, inspection, excess reviews and signoffs, poor filing systems, lack of needed reference material, excess walking, waiting, and incomplete information.

The solution was to develop a cross-functional work cell to put together the work instructions. As a result, many of the handoffs were eliminated and non-value-added steps were eliminated. A takt time was created, based on the demand for work instructions (demand was fairly easy to predict) and the time available to work on them. A key part of the solution was to take the core people doing the bulk of the work and colocate them in an open office environment. A work cell was physically created in the office so that the key functions required to produce a work instructions folder could pass the package from station to station in record time. The old office arranged people into separate functions and each person had an office with high partitions. In the new work cell, the key people had desks surrounding a round table. The job summaries move around the table from person to person in a one-piece flow. Valued-added time was calculated before and after—with striking results. Note that some non-value-added time is required, e.g., filling out paperwork that adheres to Navy policies even if it is not necessary to get the mechanics what they need. We separate this out from “wait time,” which is pure waste.