Separating Human Work From Machine Work [Infographic]

Most of the work we do today involves interactions with machines. It is true not only in manufacturing but in many other business processes. The machinist works with machining centers, the pilot with an airplane, the surgeon with a laparoscopy robot, the engineer with a variety of computer systems,…, not to mention the automatic appliances that relieve us of household chores. In fact, I think that being good at working with machines is so essential that I wrote a book about it. For the short version, see the following A3/tabloid infographic. To enlarge it, click on the picture, and then on “View full size” in the bottom right-hand corner.

Separating Human Work and Machine Work

One of the key points of TPS on this subject is the idea of separating human work from machine work. It means designing and executing the work with different timelines for the people and the machines involved. If an operation is such that the same people and machines must be present and involved from start to finish, there is only one timeline. If, on the other hand, the operator can start an automatic cycle on the machine and walk away trusting that it will complete successfully and that there is no need to return until then, you can plan for the operator to do other work in the meantime, meaning that the operator and the machine are on separate timelines, that meet only at the start and the end of the machine cycle.

We experience this in daily life with washing machines. We load them up, select a program, add a dose of detergent, press “Start” and walk away, confident that it will be done in, say, 60 minutes. The same is true of most kitchen ovens. There is no need to stand in front of these machines the whole time. They are on their own timeline, and you on yours, and they even know when to stop. In theory, the same should be true of your office printer. You should be able to tell it to print double-sided and collate 25 copies of your 50 page course handout but you can’t, can you? The minute you turn your back, it jams, or runs out of paper or ink/toner. If the printer is next to your desk, you may be able to get snippets of work done between stoppages; if it is in another room, you’re stuck standing in front of it the whole time, clearing and restarting it at random intervals, and idle in between.

Clearly, you can be more productive with a machine that actually works on a fire-and-forget mode than with one requiring baby sitting. The real question is how to get there from here, and there are two challenges to overcome for this to happen:

  1. Manufacturing or process engineering. Technically improving your processes is an engineering challenge at multiple levels, from physics and chemistry at the core to the handling of parts inside the machine and operator interface design.
  2. Planning the work or people and machines. Once you have made it practically possible to move away for the duration of an automatic machine cycle, you need to lay out the machines and organize the flow of work with the goals of keeping the operator doing useful work and making effective use of the machines.

3 comments on “Separating Human Work From Machine Work [Infographic]

  1. The concept of transferring physical and mental work from the operator to the machine was one of the first elements of what was later to develop into the Toyota Production System.
    Historically in the textile industry each loom had its own operator. They were there to change empty shuttles, watch for thread breakages and when they occurred, stop the loom and repair them. In 1924 Sakichi Toyoda designed a loom that transferred these actions to the loom, and incorporated automatic shuttle change, and a mechanism that automatically stopped the loom if a thread breakage occurred. This made it possible for one operator to run several looms, and ensured the machine stopped automatically if a defect creating condition occurred, this improved productivity and product quality. These ideas developed into the Jidoka element of TPS. —

    A licence for these features was sold by Toyoda in 1929 to a British loom manufacturer. I have read that these funds were used by Kiichiro Toyoda to establish Toyota Motors in 1935. In the late 1930’s Kiichiro started to develop Just in Time manufacturing. Jidoka and JIT became the pillars of TPS. —

    The application of Jidoka will improve productivity and quality by transferring the movement of the operator’s physical and mental activities over to machines, equipment and tools. Understanding these basic principles for achieving focused waste free flow & zero defects is the foundation of lean operations, —

    JIDOKA. PHYSICAL WORK TRANSFER. In this area we must transfer as many of the operator’s physical movements as possible to the machines, equipment and tools. The goal is to have the machine run without the supervision or the operator. One operator can now operate several processes. This is essential element for achieving the maximum operator productivity in the one piece flow cell design. —

    JIDOKA – MENTAL WORK TRANSFER. If the machine was just left to run unsupervised by the operator, it could be producing defective parts. We must therefore transfer the operator’s mental work, thinking, to the machine. This will give the machine a mechanical intelligence that will automatically stop the process if any defect creating or unsafe working conditions occur. The operator can now safely (Zero defects – Zero damage or danger) operate several different machines within one sequence cycle. This facilitates grouping machine in line with product process sequence, rather than machine type.
    In the development of TPS, the quality element of Jidoka evolved into Source inspection and the Poka-Yoke system. —

    • A couple more details on the loom story… As I heard it, the loom operators were changing bobbins within shuttles, rather than whole shuttles and, until Sakichi Toyoda, the automation efforts were also towards bobbin exchange and were unsuccessful.

      One of Sakichi Toyoda’s insights was that changing shuttles was much easier than changing bobbins within shuttles. It’s like the difference between reel tapes and cassettes. It is the kind of ideas that is retrospectively obvious, but that it took a long time for anyone to come up with.

      I also heard the story that the sale of the patent on the Type G automatic looms to Platt Brothers for 100,000 pounds financed the startup of the car business by Kiichiro Toyoda. But it is the official story, and I take all corporate official stories with a grain of salt. A car plant today is a billion-dollar investment. Could the sale of a patent for a new loom have provided enough in Japan in the 1930s? I don’t know, but I find it difficult to believe.

  2. Michel, the breakthrough on the automatic shuttle change on Toyoda’s Type G* loom, came from having a magazine containing several shuttles on the loom. When the weft thread in the existing shuttle was running down, it was automatically changed for one from the magazine. The machine automatically stops when there are no shuttles left, or the weft thread broke. I think £100,000 in 1929 would have been more than sufficient to fund the initial product development, and limited production of the first Toyota cars. I am sure Kiichiro Toyoda would have had access to the extensive manufacturing facilities within the Toyoda Loom works.

Please share your thoughts: