Technology

Karakuri doll serving tea

Sep 17 2013

What is Karakuri Kaizen?

Google “Karakuri Kaizen,” and you see a small number of Youtube videos from Japan, Thailand, Italy, and Hong Kong showcasing materials handling devices that rely on gravity, levers, cams and inertia to move bins in elaborate ways, transfer parts between machines, or deliver a controlled number of small parts to an operator’s hand.

Here is one from Japan’s JMAC with multiple examples:

Such devices have long been used as part of TPS and Lean, but now we have a generic name for them. The principles of Karakuri Kaizen given at the end of this video are as follows:

  1. Don’t use the human hand. Move objects automatically.
  2. Don’t spend money.
  3. Use the force of your equipment.
  4. Build it with the wisdom and creativity of the people of the shop floor.
  5. For safety,  don’t just rely on paying attention but build a device that stops automatically.

While “Karakuri Kaizen” is an alliteration that rolls of the tongue almost as easily as “cash for clunkers” or “toys for tots,” you may still wonder where “Karakuri” comes from and what it means. Until “Karakuri Kaizen,” I had never heard it stand-alone but always as part of “Karakuri Ningyo,” or Karakuri Dolls, which are wind-up automata with wooden gears and levers developed at toys in 18th-century Japan. The best known are tea-serving dolls, like the one in the featured image.

As Karakuri dolls are a reminder of ancient ingenuity, the term has a positive connotation in Japan. I once used a picture of one in a magazine ad for US-made automation software, to connect the product with the local culture. But the term, obviously, means nothing to anybody who is not Japanese.

By • 6 • Tags: , , , ,

Jul 23 2013

Standard Work in Low-Volume/High-Mix Manufacturing

In the TPS Principles and Practices discussion group on LinkedIn, Brian Miller initiated a discussion on “How do you create standard work for a customized product that has over a billion combinations?” It has had 31 comments so far, and I would like to share here a few that I made.

Even in a plant that is perceived to be focused on low-volume/high-mix production, you usually have an uneven demand pattern, calling for different approaches to standard work by product category.

You start with a Runner/Repeater/Stranger analysis to determine what it is we do often and what not. Without this analysis, we commingle in the same lines products made every day with other products made sporadically (See Lean Assembly). In Japan, this is called P-Q, or Product-Quantity analysis, with the categories called A, B and C. The more vivid Runner/Repeater/Stranger terminology comes from Lucas Industries in the UK. You then use a dedicated, integrated production line for each Runners, a flexible line for each family of Repeaters, and a job-shop with functional groupings of equipment for Strangers.

Then, obviously, you face different challenges for developing standard work in each category:

  1. In a runner line, you can post A3 sheets above each workstations with instructions for the purpose of allowing supervisors to monitor how the work is being done. This is the normal situation of high-volume production.The operators themselves do not need to read the instructions for every workpiece; as soon as they are proficient in the job, they work from memory.Operator instructions in auto parts assembly
  2. A repeater line is for a family of products with variants but with a high commonality of materials and processes. The instructions that can be posted on A3 sheets are then limited to the common processes, but the operator needs to read what is specific to each workpiece. For final assembly of cars, Toyota has provided “build manifests” printed on larger sheets mounted on car bodies and bearing all the option information. The kitting of workpiece-specific components also helps. In computer assembly, workpiece specifics are shown on electronic displays, with component picks validated by auto-ID technology, including bar codes, QR-codes, or RFID chips.Toyota build manifest Valenciennes cropped

    Kit-trays-with-instructions-web
    Kit trays with instructions for repeaters in electronics assembly
  3. Strangers are the odds and ends with sporadic demand, one-of-a-kind systems, or new product prototypes. For strangers, you cannot rely on operator memory or habit. For one-of-a-kind systems or prototypes, you cannot even assume you have the knowledge needed to produce workable detailed instructions. Each stranger is a job in a job shop and requires instructions on a traveller that moves with the workpiece. The traveller may be hardcopy or electronic. An electronic traveller may either be a device containing all the instructions or an ID that triggers the download of appropriate instructions at each station. Strangers are usually built by skilled craftsmen able to work directly from engineering drawings. Given the nature of strangers, however, the time required cannot be precisely known and sequencing mistakes will happen, making rework inevitable.
Stranger assembly in Yatai at Omron
Stranger assembly in Yatai at Omron

By • 10 • Tags: , , , ,

5S example from Korry

Jul 9 2013

Visual management as a “tier 2” tool

In the TPS Principles and Practice discussion group on LinkedIn, Emmanuel Jallas asked whether visual management was a “lever to make other tools work, or  a tool by itself.”

Visual management as embedded into other tools

As I see it, visual management should be part of everything else we do, but not treated as a stand-alone topic. Visual management should be considered in the design of a plant, a production line, a supermarket, a shipping/receiving area, a crossdock, a cafeteria, a restroom, etc. It is part of setup time reduction, cell design, or kanban implementation. It make visual management a “tier 2” tool.

If, however, you try to discuss it or teach it as a stand-alone, generic subject, you quickly get dragged into what it involves in different, specific contexts. If you have a mixed group, whatever you say will only be of interest to a minority at a time. On the other hand, if you are teaching Lean Logistics, then the discussion of visual management in materials handling comes naturally.

If you write a “How to” guide, you really have to think who your intended readers are. If you write on how to design a machining cell, you know exactly who they are. And, if you do a good job of writing it, all of it will be of use to this audience.

But who is the audience for visual management? It’s everybody! But the general theory of visual management fits in a few pages. After that, you have to go to examples, and each example is for an application that is only of interest to a tiny sliver of a manufacturing organization. So maybe 1% of your How-to book is of interest to each reader, but you can’t cut any of it, because another reader’s 1% is somewhere in the  remaining 99%.

By the way, people like Gwendolyn Galsworth or Michel Greif, who have written several books on visual management obviously disagree with me on this. I use their books like dictionaries, not how-to guides.

Visual management and Potemkin villages

Since visual management is, … visible, it is commonly part of the Potemkin villages put up by companies that want to look lean to outsiders. But the fakery is easy to spot when, for example, you see bins under a sign that says the area should be clear, the operators don’t know what the colors on the andon lights mean, the color codes are inconsistent across the floor, or a production monitor shows overproduction and production continues,… It does not take many discrepancies to torpedo the credibility of visual management.

Complicated color codes are a tell-tale sign that a system is not used.  The andon lights I have seen in Japan have only three colors with one and only one solidly lit at a time. It’s Red, Yellow, and Green, with Red meaning that the machine is stopped, Yellow that it is available, and Green that it is working. Used consistently throughout a shop floor, it gives you an overall equipment status at a glance.

Of course, the light suppliers prefer to sell more elaborate models, but I have never met an operator who could tell me what White with blinking Blue was supposed to mean, especially when it was not consistent across machines. So, if you see that, you know that the lights are just there for decoration.

Visual versus verbal communication

Reliance on words is not recommended for an audience that does not have a common language. That is why traffic signs in Europe are mostly wordless and European car dashboards are covered with pictograms, that are sometimes but not always self-explanatory, which is why I have taken to calling them “euroglyphs.”

An American car dashboard with words is actually easier to understand, but only if you know English. Like European roads, a California production shop floor may have a work force with multiple nationalities and uneven English proficiency. As a consequence, using words for instructions or safety warnings is not much of an option.

Two resources I find helpful is thinking through these issues are usability engineering experts Don Norman, author of The Design of Everyday Things, and Asaf Degani, author of Taming Hal.

If you include hearing, touch and smell, I suppose it should be called “sensory management” rather than “visual management.” If we use “visual management” for all forms of sensory management, what term are we going to use for what is specifically visual?

Visual management as part of the information system

The term “information system” should encompass all the means used in a plant to exchange and process information. Visible management is part of it, along all the computer applications, from CNCs, PLCs and SCADA systems to corporate servers for technical and business data. They are all components of the same information system and both are needed to run a plant.

Japanese terms for visible management

The Japanese term I have heard for visual management is ”medemirukanri”(目で見る管理), literally “management you can see with your eyes.” Mieruka (見える化) is new to me; it means “transformation into something visible.” I see it as an improvement, as it is shorter and just as self-explanatory. I suppose you could say that medemirukanri is the result you achieve and mieruka the process by which you achieve it, but I don’t see that nuance in the usage.

By • 5 • Tags: , ,

Jun 21 2013

Using videos to improve operations | Part 6 – Quick simograms

Here, we finally start collecting measurements from the video, focusing on what we can collect while watching without stopping. In this mode, we can break down operator time by broad categories like  “Waiting,” “Walking,” or “Assembling,” but we don’t have the time to name each task and collect comments or improvement ideas. This will require a more detailed and time-consuming analysis. 

One method, developed by Christophe Caberlon, involves two analysts, one viewing the video and the other one filling out an electronic spreadsheet. Instead of looking for state-change events in the video, we look at it in 5-second increments. Every five seconds, the analyst viewing the video calls out the state the operator has been in since the previous call. Each 5-second. Interval is assigned one column in the spreadsheet and there is one row for each state. Based on the call, the second analyst switches the color of the cell for the state and time interval.

Counting in 5-second intervals involves aliasing, but it is not a problem for a rough-cut estimate. The rows in the spreadsheet do show the state transitions in a Gantt-chart like format called “simogram,” and can summarized into proportions of time spent in each state, as in the following example:

Simogram example

This example uses cell background color to express content, which is not generally recommended because Excel does not provide built-in tools either for quick input or for analysis. The result, however, is graphically much more attractive than filling the cells with Xs. Changing the background color of a cell requires multiple steps, which cannot be repeated every five seconds. These steps, however can be recorded as a Macro. In this example, the macro has Ctrl+q as a hot key to mark a cell and Ctrl+w to unmark it. Also, each 5-second time segment must be assigned to one and only one category. When working your way through a video, it is impossible to avoid cases where one segment will be missed and another accidentally assigned to more than one category.

To detect these errors, we need to count the gray cells by column, and to summarize the times into relevant aggregates, we need to count them by rows. While Excel provides no built-in function to do this, you can find add-on modules to do it. The modules used above are due to C. Pearson. 

This method is also restricted in the number of states to track. It is feasible for two or three but not fifteen. With the limited number of choices, it is a good idea to include an “Other” state. The states should also be clear and unambiguous, such as:

  • Walking: the operator’s legs are moving.
  • Working: the operator’s hands are moving.
  • Waiting: all the operator’s limbs are still.
  • Touching: One of the operator’s hands is touching the product.

Categories that are abstract and subject to interpretation, like “Value-added” should be avoided. Note also that an operator who is Working or Touching, may be handling the work piece or transforming it, and we don’t have enough categories at this level to make the difference. 

Timer Pro provides a method called “Non-stop timing,” in which the analyst simply clicks on a category when observing a state transition, and the time since the previous click is automatically assigned to this category. This eliminates the aliasing due to using 5-second intervals, and relieves one analyst from the task of clicking the right spreadsheet cell every 5 seconds.

By • 2 • Tags: , , , ,

Jun 8 2013

Using videos to improve operations | Part 5 – Watch it in fast motion

Christophe Caberlon has a special way of conducting this first step: at 6X normal speed. When reviewing videos with the operators who have been recorded, he has found that accelerated motion has the effect of drawing attention to the activities that consume the most time, as the ones that do not are filtered out. If the operator spends the majority of the time walking back and forth to a shelf to pick parts, a fast-motion viewing of the recording makes it dramatically obvious and compelling, making the operator eager to help improve the process. After trying various 2X, 4X and 8X, he has found that the 6X ratio of acceleration works best for this purpose.

When Christophe  told me about this, it reminded me of a scene in the movie The Hunt for Red October, in which sound is used the same way. Jonesy, the sonar operator in the American submarine tracking the soviet submaring Red October, has just heard the sound of the propellers disappearing when the Russian boat turned on its “silent” propulsion system. But he still heard a vague rumbling. The analysis software categorizes this sound as magma displacement, but the sonar operator believes that the sound is made by the submarine. To convince his captain of that, he plays the recording at 10 times the speed, which turns the rumbling into a rythmic “tac tac tac” characteristic of a human artifact and not magma displacement.

The idea of watching the video in fast  motion is that, by filtering, it lets you see patterns you would otherwise miss. The same is true of slow motion, but fast motion has the advantage,… of being fast. You wouldn’t want to review an entire 1-hour video in slow motion, but you may do it in a few excerpts as part of the detailed analysis we will go into in a coming post.

Of course, the idea of reviewing a video in fast motion is more interesting if you can actually do it, and the most common video players around don’t let you run at 6X. I tried the following:

  • Quicktime V. 7.7 through “A/V controls” lets you run videos at speeds up to 3X.
  • Windows Media Player does not let you change the speed.
  • Real Player lets you change speed by powers of two by pressing Ctrl-Shift-F4, but the resulting accelerated videos are not watchable.
  • VLC media player lets you use any speed up to 4X, and powers of two beyond, up to 32X.

Among video editors, Windows Live Movie Maker lets you change speeds, but Google’s Picasa 3 doesn’t. For Windows Live Movie Maker, it is again in powers of 2.

In video annotation software, ANVIL doesn’t do fast-motion, but Timer Pro does. To view a video in 6X in Timer Pro, you click on the Video tab and set the speed to 12. The video then plays at the right speed but the a wrong aspect ratio. If, however, you click on the Comparison tab, you can view at 6X normal speed and with the correct aspect ratio, albeit in a small window that you cannot adjust.

By • 1 • Tags: , , , , , , ,

«‹ 14 15 16 17›»