< back

Your GUI in any color, so long as it’s black

October 20, 2016

Your GUI in any color, so long as it’s black

by Francisco Almada Lobo, CEO of Critical Manufacturing
The following article was published on October 17, 2016 on LinkedIn Pulse series.

Ford Model T and mass production

Long gone are the days when Henri Ford said about his Ford Model T that "any customer can have a car painted any color that he wants, so long as it is black."

It’s important to provide the right context for this sentence though. The Model T became the world's first affordable car. It was mass produced on a moving assembly line, and was being manufactured at a rate of almost 10000 cars a day.

At that time, more than half of the cars available were Model T’s and each car was being built in less than two hours. With the demand the car was having, any minute gained would have a dramatic impact and Ford figured out that black paint dried the fastest.

Graphical User Interfaces in manufacturing

Although a bit farfetched, Graphical User Interfaces were all black painted for similar reasons. The first GUI’s were “expensive” to built. The ones long enough in the industry to have tried Motif or Visual C++ MFC, just to mention two examples, know what I am talking about.

But when applied to manufacturing, the problem gained a new dimension. On one side, these solutions simply can’t fail. Wrong inputs or any other malfunction can have really serious implications. So one can never be too careful about it and GUIs need to be carefully thought and tested, not only in development environments, but also in real production mode (this is btw, a very important aspect often neglected when evaluating manufacturing software solutions in general, and MES in particular: such a solution needs to have a sound quality system behind the development and deployment activities - that’s where CMMI comes into place).

But more than that, in manufacturing we always had the need to create interfaces which are simple to use and learn. And that meant standardization of GUIs. If GUIs all looked the same, new employees could be easily trained. If operators needed to change from one area to another, the same GUI would make their new job much easier.

The greatest common divisor or the least common multiple

The drawback? Using a mathematical metaphor, they could be:

a. The greatest common divisor - meaning they’d have only the functions that were common to all the manufacturing areas - this would leave behind specific functions that’d need to be performed in other applications, or having data recorded in paper or excel sheets.

b. The least common multiple - meaning the interfaces would have all needed functions for all manufacturing areas - this would mean several blocks of the GUI with no meaning or no usability within specific areas.

So this was an area of deep investigations and conceptualization within our company. Like any other MES provider (at least the ones who do care about usability), we tried to make our interfaces simple, yet informative, and simultaneously intuitive and easy to learn, yet allowing complex operations to be performed.

But we soon realized there was another angle too look at in MES GUIs, particularly for the functions which need to be performed by shop-floor personnel: there’s always specific requirements the GUI needs to fulfill. Sometimes this is caused by real hard requirements; some other times it is caused by the way people are used to operate; or even some other times it’s just the specific look and feel of the application to match the corporate guidelines.

No matter what the reasons were, we ended up having to customize, modify, enhance or change existing interfaces. Now, I have to say, our application made this easier than any other I had seen: it was possible to configure many aspects based on security/permissions; it allowed adding specific actions and create add-ins on existing interfaces; and it even allowed creating new GUIs based on the framework, leveraging lower level functionality and look & feel.

The shop-floor GUIs of future

Still it was not enough. We wanted a solution that would allow manufacturers to create their own interfaces. These would be made by common MES elements with domain behavior that could be set, organized and customized. Then companies would be able to create facility, area, or even equipment or role based GUIs. MES covers so many functions these days, whether it’s operations, quality, maintenance, logistics or engineering functions. So the user interface must be flexible and adaptable to whatever the role of the person using it is.

But not just that. These interfaces would have to be created without code. Because we believe these must be done by end users. By the process owners, not by programmers.

And finally the devices. Shop-floor solutions must be available for desktops/laptops, but also, and increasingly so, for mobile devices, whether these are tablets, phablets or mobile phones. So modern GUIs need to provide more than lip service to mobile. Interfaces need to be conceptualized having in mind the different usages.

So we decided to do it with Critical Manufacturing 5.0. Using Google’s Angular2 and HTML5 interfaces we created a solution allowing companies to create their own interfaces with significant productivity gains. Here’s a short video explaining it.

More info can be found at: http://www.criticalmanufacturing.com/en/critical-manufacturing-mes/diy-gui

< back


IIoT Has a "Thing" for MES. Why IoT Platforms Won’t Replace MES for Industry 4.0
by Iyno Advisors and Critical Manufacturing