AI - Obstacles

The currently still biggest obstacle in the implementation of digitisation technology is an often predominant irritation of the decision-makers responsible for production as to how the project should be approached at all and what time or monetary expenditure is involved with which measurable advantage.  The question arises of the general and temporal amortization capability of such a production optimization system. In numerous media channels, fears are stirred up and hopes are spread without being able to make concrete promises of benefits. How do you manage to keep a clear head?

In the context of this orientation article we want to list the relevant fears and possible obstacles and evaluate them if possible.

The basic opinion of most production companies, also due to their obligations under the company's confidentiality and data protection regulations, is that, as a matter of principle, all production data and information must not be passed on to third parties or to external data systems (cloud, external IT services, Internet) in order to protect against loss of know-how. 

In addition, it must be ensured at all times that installed operating systems are protected as far as possible against unauthorized access from "outside" ("hacking"). For example, WLAN-capable sensors and devices may only work with encrypted data, so that this data cannot be accessed and used without permission. The user usually wants a practical, robust and secure data acquisition system as an in-house solution. WLAN-enabled sensors and devices are often not sufficiently usable in production systems due to the high strength of electromagnetic fields. In addition, WLAN-enabled sensors often have to be equipped with a power supply so that networking through this power line would already be possible without transporting signals over a relatively insecure WLAN network.

Cloud-based solutions promise the optimal use of artificial intelligence and secure data storage. But the decision-makers responsible for production are shying away from shifting their production experiences to the Internet because it is perceived as an unsafe place. The in-house solution appears to be the most secure solution at the moment.

Another major obstacle is the lack of availability of qualified personnel. Small and medium-sized companies in particular cannot rely on centralized IT departments that are able to deal with the rapid implementation of the digitization issue. There is a lack of time and technical expertise to address the topic in-house. In this context, however, the use of consulting firms does not appear to be the most appropriate solution. What is needed are craft-oriented companies that offer a fast and reliable implementation and keep the financial risk of the production company as low as possible. This can be ensured by allowing the production company to test and experience a data acquisition system free of charge before deciding to purchase one. In most cases, useful lives of up to 6 months are recommended in order to build up a serious basis of experience.

However, not only the hardware is needed, but also the industry-related expertise to define and monitor limit values, to keep error logs and to make them efficiently usable, instead of despairing of too many error messages.

Further obstacles are the legal issues. Who is liable for production losses if algorithms do not work efficiently and effectively, if the optimization goal is not achieved in the desired or agreed time? The legal bases are not yet sufficiently prepared for these developments due to the lack of experience. Can sales promises made within the framework of a digitization project for autonomous production be fulfilled at all? Unfortunately, the answers to these questions must first be experienced and experienced for oneself. There is almost no supplier of a digitization solution who can define a concrete value proposition as an assessable performance characteristic. Such a fuzziness of project success statements and thus concrete project risks cannot be clearly eliminated in the run-up to a digitization project and therefore remain throughout the course of the project as long as no positive experiences are made. This therefore requires a certain degree of willingness to take risks on the part of the decision-makers responsible for production. Since decisions are only made in groups and rarely by one person, delays in implementation occur. The market virtually blocks itself. There are no insurances or warranty promises. So the only "first mover" is the courageous, risk-taking entrepreneurial type who takes responsibility and invests the money in an uncertain outcome? Not quite, at least up to the performance level of limit value monitoring for clean predictive maintenance (see Chapter 2), reliable promises can be made to provide additional protection for the system. Performance promises become diffuse only from the moment that software-based mathematical modelling is used as a decision support.

This means that the recommendation already made in Chapter 1 remains valid.

Data should first be collected, analysed, observed and understood in real time before it can be evaluated or autonomously controlled by ML! Potential benefits and added values result from the handling and the empirical values in the application. They can almost never be predicted completely or with complete accuracy. The way becomes the goal - just like in real life.

As a positive example, however, it is worth mentioning that the possible installation of such a digitalized production monitoring system can significantly reduce the reject rate (e.g. reduction of thermally damaged components) as well as production downtimes by means of a predictive maintenance approach that precedes the use of artificial intelligence. The scrap costs are usually much higher than the total investment in a suitable production monitoring system with possible AI connection.

This would significantly reduce the risk of whether the use of an artificial intelligence defined as a long-term goal generates measurable benefits within the production control system, since the payback can be achieved in the early project phase of predictable maintenance. All possible production advantages that arise in a later project phase would thus be seen as an additional increase in profitability or improvement of the selected, company-specific optimization parameter. The "Predictive Maintenance Approach", which can now be unerringly implemented using the data acquisition system described above, would thus serve as the economic legitimation for the start of a project to generate artificial intelligence in production systems.

In this article relevant products:

These articles from our magazine may also interest you:

Advantages of Coolant Monitoring Systems

Coolant monitoring systems can contribute significantly to increasing reliability and efficiency by collecting and evaluating various measurement and status data in the cooling lubricant system.

Click here for the article

AI - How do you proceed?

The first step in context is to start the implementation as soon as possible and not to wait for a competitor to come up with a solution. Because this solution, if it would be accessible at all, can certainly not be simply copied or purchased at the push of a button.

Click here for the article

AI - Risks

When using artificial intelligence in production, there are certainly risks with regard to a possible malfunction of the "artificial intelligence" and resulting losses in productivity and profitability.

Click here for the article

AI - Challenge

The challenge of using artificial intelligence in production engineering is to generate data whose values are as correct as possible (good data).

Click here for the article