Visual Management

[Featured image from Edward Tufte’s Envisioning Information)

Tracey Richardson started a spirited discussion on LinkedIn just by posting this quote, presumably from Toyota training materials, and asking “what else would you add to describe ‘Visual Management’”:

About Tracey’s Quote

While this quote certainly has a context, Tracey was asking readers to respond to it as a stand-alone statement — that is, out of context. From wording like “making all standards, targets, and actual conditions highly visible in the workplace,” you can’t escape the conclusion that the quote presents Visual Management as strictly communication. Tracey, who has worked 20+ years at Toyota, elaborated on the management aspect:

“Visual Management has a micro / macro process of ‘thinking’ that supports people development / leadership actions / KPI usage and standards created vertically and horizontally that align to purpose and value for long term growth and sustainability as an organization. It was a key for me in my experience.”

Defining Visual Management

The level of abstraction in Tracey’s quote is necessary when you want to encompass Visual Management in any kind of activity. If you focus on a production shop floor, you can be more specific:

Visual management is action on the shop floor based on features of objects that make their function, status, and expected performance understandable on sight.

In this context, it includes andon lights, signage, markings, posted instruction sheets or screens and charts posted or displayed. “Visual” is actually over-restrictive because other senses are involved too. In some factories, shop-specific tunes are used over the public address system to signal breaks and the location of malfunctions.

A light going red is not the only signal to operators that a machine needs attention. They are also on the lookout for unusual rattling sounds or odors, or the feel of a surface to the touch. Just as reducing clutter enhances visibility, reducing background noise makes abnormal noises easier to hear. Visual communication only becomes management if it’s heeded. A sign that says to keep an area clear is not visible management if junk piles up under it.

Gwendolyn Galsworth’s View of Visual Management

Gwendolyn Galsworth distinguishes between a Visual Workplace and Visible Management:

[…]the difference between the visual workplace and visual management is telling and important. It is also a functional difference.

These differences are not a matter of personal preferences (as some remarks seem to suggest). The differences are substantive by design. For example, you can segment visual devices by purpose and by effectiveness–their power to get us to do the right thing.

But not just that. Other dimensions of meaning as well. This segmentation allows us to dig deeper into the informational landscapes of our companies. As a result, we develop a language of visual solutions that goes so much further than mere normal/abnormal distinctions.

Stories About Visibility

The presence or absence of Visual Management in operations is best understood through examples. Here are a few that I have personally experienced over the years. The stories are real but the illustrations are recreations.

Locating people in a hoteling office suite.

A consulting firm that coached clients in visible management had a hoteling system in its head office. The consultants mainly were offsite and had no permanent desks. On-site, they each had a cubby hole for their papers and grabbed an available seat at a table in any one of a number of rooms in the suite.

Neither the rooms nor the desks had IDs. To find anyone, you had to scan all the rooms. I suggested having a board with a map in the lobby, with a magnet for each member to mark their location. The suggestion was turned down because “the board would have been ugly.” This is what it would have looked like:

This raises the question of whether there truly is a trade-off between beauty and function. Are there any circumstances where it makes sense for a business to make work harder for employees in order to preserve appearances? By the principle of functionalist architecture, form follows function, and workplaces should be designed strictly for their function. It even morphs into the aesthetic criterion that the beauty of an object lies in its fitness for its function.

Regarding a workplace, however, we need to consider what its function actually is.  A bank branch looking like the best manufacturing shop floor would not inspire confidence in customers. Projecting an aura of established wealth is part of its function and must be a consideration in its design; on a manufacturing shop floor, it’s all about getting the product out. The leaders of this consulting firm thought their head office should look more like a bank branch than a manufacturing floor.

Andon Lights on Injection Molding Machines

A row of injection molding machines had long setup times that kept them from making the right mix of products every day. Rather than start a SMED project, however, the manager chose to place andon lights on each machine. It was easier to do and an “improvement” you could show visitors. It did not enable the shop to meet the daily demand.

In this case, Visual Management was simply the wrong priority. Andon lights would have been a useful addition once changeovers were sufficiently quick but were no substitute for them.

Inventory Accuracy in Open Warehouse

The inventory in a small components warehouse was inaccurate because operators helped themselves in it without recording what they took, particularly on swing and graveyard shifts. They could have locked the warehouse and used a clerk to dispense the components, but it would have been expensive and would have soured the relationship between Production and Materials. First, they posted a memo from the Materials Manager in small print with a stern warning, which had no effect, as nobody read it. Then they replaced it with a sign patterned after those used in local traffic: a pictogram in a red circle with a slash through it. The pictogram showed a figure grabbing parts from shelves. Above it was a one-line caption wordsmithed to get the message across without offending. It started a conversation.

This was a case of struggling to find the lightest touch to achieve a result. It was based on the assumption that the operators did nothing more than resolve shortages on the line in good faith, with no intention of hurting inventory accuracy. This action was, of course, no substitute for eliminating shortages but it could be implemented right away.

Visual Management as a Subtopic

Visual Management overlaps with many other disciplines, from user interface design for production machinery to 5S,  production control, maintenance, quality, daily management… In fact, anything you may want to improve in the operations of a factory has a dimension of Visual Management.

Gwendolyn Galsworth’s perspective is that visuality is the umbrella over everything, including mistake-proofing. It is intellectually challenging when you consider that most mistake-proofing devices are not visual. Just making you see that an action is a mistake does not prevent from you making it.

Mistake-proofing devices either physically prevent you from making a mistake or from passing on a misprocessed workpiece. A tang that keeps an ink cartridge from fitting in the slots for other colors does not rely on sight. Neither does an air jet to blow an empty carton off a packaging line.

Rather than a main topic, I believe in treating Visual Management as a subtopic in every project. SMED, cell design, warehouse operations, supply chain management,… all have an application-specific component of Visual Management. There is just not enough meat to generic Visual Management.

The Blurry Boundaries of Visual Management

The boundaries of what qualifies as Visual Management are blurry. The signals you can see on the shop floor with your naked eyes — that others can see as well — are part of it but what if you are the only one seeing them through Augmented Reality (AR) glasses? A manager reviewing a performance chart in the office is not doing Visual Management. If the manager posts the same chart on a performance board on the shop floor, does it become Visual Management? What if it is at such a high level of aggregation that no one on the floor can relate it to their daily work?

People can see, hear, touch, smell, and taste — often better than sensors — and leveraging these unique capabilities in operations is part of showing respect for their humanity. Yet, among all the senses, sight gets special treatment. Visual Management is a topic but the use of any other sense is only mentioned in passing. In touch quality control, for example, an assembler runs a hand over each workpiece for tactile feedback on the presence of all the components mounted at the previous station.

We value the use of sight more. It provides the most information fastest and data in visual form has more permanence. Signs stay in place until taken down but visual signals can be as transient as a one-time knock in a machine, as in the case of a lookout spotting an enemy ship:

Sources on Visual Management

Visual Management as we know it originated in Japan. Therefore, we should check out the Japanese literature on it. European and American researchers, however, have also made many contributions in this field, even recently.

Japanese Sources

In his 1978 book Toyota Production System, on page 129, Taiichi Ohno has on paragraph about “Visual Controls” or “Management by Sight.” In the original, he called it “me de miru kanri” (目で見る管理). Google now translates it to “Visible Management.” If, however, you just translate “me de miru” (目で見る), it comes out as “see with the eyes” or “see with one’s own eyes,” not “visible” or “visual.” Here again, it seems that nuance is lost in translation. It’s management based on what you see with your own eyes, on the shop floor, unmediated.

Charts on a performance board are curated data. They are, however, a large part of what we call Visual Management today. In 1993, Nikkankogyo published a “Big Dictionary of Visual Management” (目で見る管理大事典), with a  cartoon summarizing what it looks like. Here it is, with the captions translated:

The “TPM” board on the right contains charts that summarize things that you don’t see. The caption on top is odd. TPM is supposed to be focused on equipment, not product quality, which is the subject of the other board. The rest of the book has many examples of communication devices and A3 templates. They cover communication on various subjects but is short on what to do with them. This picture is also yet another example of using manga to communicate with an audience of manga readers.

American and European Sources

For ideas on this topic, you find good sources in the US and Europe. When writing Part I of Working with Machines, I drew on Don Norman’s ideas on usability engineering in The Design of Everyday Things. It is about designing machine interfaces that are intuitive, easy to learn, and not error-prone. In his research at NASA, Asaf Degani took it one step further. He was working on designing airliner cockpits to produce messages that pilots would interpret correctly, which is key to preventing crashes in case of emergency and reported his work in Taming HAL.

For designing performance boards or A3 instruction sheets for operators Edward Tufte’s work is a gold mine. Jean-Luc Doumont, although focused on scientific communication, also has useful ideas, particularly on the informative captioning of charts.

In addition, print and online media often use combinations of text and pictures called infographics whose design can inspire and inform shop floor visualizations. For inspiration, you can browse collections like Best Infographics, or you can take guidance from infographics authors like Alberto Cairo. If you want to go back to the origins of the concepts used in infographics, check out the isotypes of Otto and Marie Neurath from the 1930s in International Picture Language.

Conclusions

The manufacturing world has been slow to adopt Visual Management. Good ideas have been around since at least the 1930s, and advances in IT have only made these ideas easier to implement. It requires paying attention to the engineering of human work, which Toyota has done with TPS but few others have. The key question going forward is whether Visual Management is a point of intervention, as Gwendolyn Galsworth has been arguing, or an item within the list of dimensions to be addressed in every single improvement project, which I believe is more effective.

#visualmanagement, #usabilityengineering, #isotypes, #traceyrichardson, #gwendolyngalsworth, #tufte, #donnorman, #ottoneurath, #marieneurath