How to approach IoT and Industry 4.0?

Some time ago, the managing director of a medium-sized company in southern Germany spontaneously asked me: "How should we best approach the topic of IoT/Industry 4.0 in our company?" Not in a business context, and since I also had a private contact with the questioner around two corners, I was - also as Sales Director IoT - very anxious for a neutral answer.

I did not want to develop him into a lead out of consideration for the private connection. I didn't know much about the company itself: There was a well-functioning SAP ERP and, according to my interlocutor, the workflows and processes were so far cleanly orchestrated. And now they wanted to tackle the topic of IoT/Industry 4.0 - but how?

A good place to start is probably to first become aware of your own motivations. Is it the fear of falling behind because competitors are already further ahead in this area? Is it the (intrinsic) enthusiasm for technology and new possibilities among individual employees? As good as external impulses can be, it is still recommended to first look at one's own company and become aware of one's own strengths and weaknesses.

IoT is characterized in particular by the fact that processes can be optimized and automated, i.e. made smart, by means of technology. So we should ask ourselves: In which processes do we currently have the greatest efficiency losses, the most serious deficiencies?

In short: Where is the shoe pinching the most in your own company? Where can improvements be achieved by means of IoT? Whether the data is generated via networked sensor solutions or read out from controllers is of secondary importance, the technology anyway. Away from buzzwords and towards real added value.

The second step is to take your ideas and approaches to someone who has sufficient implementation experience with IoT to subject them to a feasibility analysis; someone who has some overview of what works (and what doesn't) empirically. It is further advisable to establish a complexity order in order to be able to estimate, evaluate and compare the efforts associated with the respective concept.

Start small and learn

On the basis of the resulting roadmap, you can then make a well-founded decision as to whether you want to tackle an IoT/Industry 4.0 project on your own or with a partner, taking into account your own resource availability and utilization. To start with, you should choose a small, agile project - and learn from it. The next steps can then be defined and implemented.

Condition monitoring is often chosen as an entry scenario, i.e. the condition monitoring either of machines and systems in the company's own production or of products used by the customer. In this way, the relevant data can be obtained, the project can be readjusted, evaluated, and supplemented if necessary. The more data that accumulates, the better it is possible to identify correlations, for example between certain operating parameters and failures.

Once sufficient data has been collected, it makes sense to look into algorithms or artificial intelligence, respectively, so that the data does not only give you retrospective correlations, but also allows you to predict (probable) developments in advance.

In addition, the technical data collected to date can begin to be put into a business context by combining it with ERP data. This is how new services and business models can be developed. That was more or less my answer. I am still in contact with my interlocutor, and now also professionally. I probably can't quite get out of my own skin after all.