Why some kind of digital transformation and automation will lead us nowhere and will bring no benefit at all!
Dirk Fischer
Chief Operating Officer at Huf Group | Passionate about transforming organizations towards holistic business excellence, based on the principles of the Toyota Way, Theory of Constraints and other effective approaches.
Precaution: This article has to be seen in the environment of manufacturing operations!
I have written already a post about this subject, but I could not resist to write an article as well. There was a great comparison of 2 pit stops posted some days ago.
Watch the video below:
Usually the Formula1 pit stops are used to open people´s eyes in order to understand how fast specific changes on equipment (setup) could be executed. Of course, people always refer to SMED (Single Minutes exchange of dies) as the core method to increase speed of change over.
I am so thankful for this video, because it is also an eye opener for some other interesting aspects.
The 1981Team!
Please do not feel attacked, but a lot of companies are only operating on the 1981 level compared to the possible 2019 level shown here.
However, many companies are currently being tempted to solve the problem as follows:
Let´s imagine that some Industrie4.0 consultant does a time warp and would provide the 1981 team the following recommendations:
1). You need a cloud to store all your data and you will need to store much more data in order to do better analytics. As we now need to handle big data, you also need to adapt your entire IT-System accordingly.
2). We need to put sensors in each object, so that we get more online data. So, we need to develop and install tire sensors, which allow us to do online condition monitoring. So we will better know the condition of the tire.
3). We then of course need the software, which will enable us to actually gather the data on line and we can store it in the cloud. This will be a kind of track and trace software.
4). The next software we need is a better forecasting software. This software will be a more accurate forecast to anticipate, when the driver will come in for pit stop.
5). Every tire and every tool needs RFID and barcode, so that they can be tracked and traced and you can look at the actual data of utilization. This will generate savings as we may found out, that some objects will not be needed.
6). We need to have a deep look in how to bring in robotics. The process is far to manual. This will also make the process more efficient (crew size can be reduced)
7). Every pit crew member needs a smart device to get ALL data whenever they want to see it.
8). The crew needs VR glasses so that they the train the pit stop in virtual reality.
Recommended by LinkedIn
The problem is, that all this technology will not improve the performance of the crew at all.
The 2019 team
So where is the true improvement coming from:
1). Detailed analysis and focus on the ACTUAL WORK
2). Found creative solutions how to improve the ACTUAL WORK
3). Improve collaboration and synchronization of the crew (work team)
4). Invent technology, that helps doing the ACTUAL WORK and supports the operator.
5). Define new standards, review them and always find better solutions.
6). Training, Training and Training!
7). Training, Training and Training!
9). Training, Training and Training!
By the way, all the great ideas from the industrie4.0 consultant will also cause no improvement at all for the 2019 team.
One of the best improvement strategies (and will always be) is to master the actual work. This is not easy, requires great (lean) leadership, physical, emotional, teaching, learning and excercizing effort and consistency. There is no shortcut.
So please be skeptical of the rain makers and keep your brain switched on!
I frequently see industry4.0 papers, where huge productivity improvements are promised. In some areas, this is maybe a little bit true, but in many areas you will gain 0.
www.gaoexperts.com
Business Development Representative @ Kinaxis | Top Performer, Supply Chain Planning
2yIt's not so much about the technology but about the technique. Supply chain planning and governance has been too siloed in discrete functions to cope with rapid change. The fundamental flaw behind this sequential and insular and modular approach is that it creates fog between functions, and nothing moves fast in the fog. You also can't trust what you can't see ... A modern business needs a real-time view of data and the ability to act on it without delay.
Strategy, Organizational Excellence, i4.0 & Entrepreneurship | MBA
3y👌
Promotor van werk met waarde voor de wereld en waardevol voor klanten & medewerkers
4yThanks Dirk for this critical view on Industry 4.0 solutions. With respect to this rather detailed fysicall activity, changing tires, I also do not see too much (not to say none) adding value of bigdata-like solutions. But thinking of all the Value Stream Maps I made myself, based on go to gemba one-off snap-shots, I tend to think that big data analysis and visualisation could be of great help to "learn to see" upfront and to provide evidence of result afterwards and to support sustaining the solutions to visualise the work in progress realtime. Anxious to hear from you what your thoughts are on that?
Vice President Industrial Engineering at Huf Group
4yLove this
Accredited ReflACT Professional | Progress-Oriented Coach | Helping Unlock Growth and Maximize Energy
4yGood article Dirk! I fully agree with your statement "The problem is, that all this technology will not improve the performance of the crew at all." I know a company that has been monitoring 30k parameters for half a year and they don't manage to fit a good surrogate model to steer the factory. And that's not going to work either, because people subconsciously assume that all important parameters are included in their list. It's based on EXISTING knowledge. The biggest problem is that people do not see that it is especially important to systematically flush out knowledge GAPS about the functioning of systems. This requires skills with regard to detuctive problem solving to uncover rich causal explanations for deviations. Big data and surrogate models are based on inductive reasoning and consequently very uncertain. There is no math to prove causality! Big Data, Big Disappointment? I think it's very likely!