Marginal Column

Prof. Alexander Verl, Dr. Eng. Dr. h. c. mult.

has been Executive Vice President for Technology Marketing and Business Models at the Fraunhofer Society since April 2014. He previously headed up the Fraunhofer Institute for Manufacturing Engineering and Automation and occupied a chair of control technology for machine tools and manufacturing equipment at the University of Stuttgart, Germany.

Photo: Prof. Alexander Verl

“People won’t be relinquishing control”

July 2014

 

Prof. Alexander Verl on the significance of automation, expertise and working ethics when accelerating the production process in mechanical engineering.

 
 

What does the market expect of mechanical engineering in regard to times to market?

The long-term objective in manufacturing is personalized mass production. Dealing with complexity is a major topic here. Procedures, machines, user interfaces are becoming ever more demanding but, in fact, it should be possible to operate the machine with only two to three commands. In terms of control technology, we should be driven by the pursuit of more elegant solutions. The essence of mechanical engineering is to implement the customer’s specifications in a way that is as precise, user-friendly and economical as possible.

That is why it all boils down to achieving a degree of automation that is so high that the customer can configure the product he desires, without the help of an expert, integrate that item directly into the production process and, while doing so, have complete control of the technology and the costs. Some carmakers are already preparing apps with which customers can configure their own dream vehicle. Finally, the app cites a price and the customer can take all this to the dealer. This is also to be possible for machinery in the future. Mechanical engineering companies will have to respond to this wish, expressed by the customers. To do so, they have to make configurators available and modularize their products. If, in addition to individual configuration capabilities, the system provides information on how the product can be automatically manufactured, then this accelerates the entire product creation process.

How can simulation and networking help to shorten this period?

Manufacturers of automation equipment have long had networking and self-organizing abilities. The task now requiring great diligence is to make use of these networks. The final stage of “Industry 4.0” will not mean that every device speaks with every other device and that hierarchies in automation will have disappeared. Those who find the right balance between self-configuration, networking, and providing production data will enjoy considerable advantages.

Simulation technology will help in the entire process. In the ideal case, salespeople will use simulation when talking with the customer to clarify whether they have actually grasped the customer’s ideas. In addition, the best components – those already available – can be selected, saving development engineers unnecessary effort. The developer, in turn, can use simulation to ensure that manufacturing specifications comply with the customer’s requests. If the simulation is accurate enough, then it will behave exactly like the ultimate machine. If the controls are trouble-free in the simulation, they will be equally trouble-free in the real-world machine.

In the meantime, many machines have become so complex that they can no longer be put into operation profitably without simulation. Using simulations can also be continued through to machine commissioning and, in fact, during later maintenance operations. If an exact model of the customer’s application is at hand, then the service department, working from a remote location, can analyze very exactly any problems that might arise or the customer’s wishes for changes.

What part will the human dimension play here, one example being the cooperation between the customer and the vendor?

During the process of acceleration, a great deal will depend upon the extent to which the customer gives us access to data. Today, software can make machines almost fail-proof. The software would, however, also record very precisely how the human actually uses the machine. This is, of course, a touchy point.

During discussions with the customer, expertise plays an important role. This is true above all when the ideal situation in simulation has actually been achieved and where sales work can be “automated”. An interesting study in this respect has been published by the International Federation of Robotics*. It shows that in countries where robotics has been used to achieve a particularly high degree of automation, unemployment is especially low.

This is because when a nation has already achieved a high degree of automation, wage costs play an ever less important role. Manufacturers who in the past had put more faith in low wages cannot simply jump on the band wagon, because the investment levels are too high. Armed with this knowledge, sales can be optimized by digital means, using appropriate software tools, so that other manufacturers would first have to invest heavily to catch up.

How do you see the course of development in this field? Where will we be in five years?

There will be a convergence at the international level, but in Asia the topics of networking and “batch size 1” will not be of much interest in the coming five years. This is because employee fluctuation in Asia is far too high. Expertise drains away continuously and new employees have to be taught. The problem is that government-planned growth in high-tech fields doesn’t work all that well. You can see an example of that in Germany, where the planned “transition to green energy” is not evolving as originally envisioned. As regards mechanical engineering, we will continue to enjoy a lead in Europe although China might have already overtaken us in other industries.

Are there new technologies that will be revolutionizing the topic in the future?

One can quite sensibly use augmented reality for this purpose. At our institute, we make videos of people as they do their work and use those videos to create avatars, which do the same work in a virtual environment. In this way we can determine physical burdens very exactly and optimize working procedures. Powered exoskeletons will also be a subject for the future. This might be the case at a smaller scale, by helping a nurse to move properly when lifting and shifting patients. But cognition – or artificial intelligence – will also appear on the scene at some future moment. At present, however, it is impossible to foresee whether we humans will be willing to relinquish control.

* “Positive Impact of Industrial Robots on Employment”, International Federation of Robotics, London 2011