Machine Learning - the future for mechanical engineering

Machine Learning - the future for mechanical engineering

Release Date: 28 February 2019
The industrial world is in a constant state of change. Machine learning will change mechanical engineering and thus many user industries. Implementation has already begun - now the focus is on concrete application scenarios and their implementation.

Machine Learning brings many new and exciting approaches, especially for mechanical engineering. The efficiency, flexibility and quality of the systems can be significantly improved with the help of the available data. New business models for customers are developed. Machine Learning ensures that software and information technology are increasingly becoming the key drivers of innovation in mechanical engineering.

In many industries, the increasing interchangeability of individual machines will mean that in future not only the machine itself will be sold, but above all supplementary services. It also explains why machine learning is on the agenda in management and in many specialist areas of mechanical engineering companies.

Where does the technology come from? Machine learning is an important part of computer science and artificial intelligence. Computer programs based on Machine Learning (ML) can use algorithms to independently find solutions to new and unknown problems. The artificial system "recognizes" patterns and laws in the learning data it receives. Tools already established on the market help to find the algorithms. "New frameworks and platforms support the broad application of these topics in everyday project work", explains Guido Reimann from VDMA Software und Digitalisierung.

The technology offers undreamt-of possibilities for machine and plant construction: Existing business and production processes can be optimized. The machines become intelligent and almost self-sufficient process service providers. Dr. Alexander Wunderle, data analyst of the transmission specialist Wittenstein, confirmed: "We can use machine learning to develop new products."

For the implementation of Machine Learning, the following questions arise again and again: How do I start such a machine Learning project? Which application scenario is suitable for my company, which experts do I need and which prerequisites should be created in order to successfully implement a project. Benedikt Buer of HOMAG Plattenaufteiltechnik is one of the developers of intelliGuide, the intelligent assistance system from the manufacturer of woodworking machines. The software architect can report from his own experience and gives a valuable tip to companies who want to deal with machine learning: "Start with a small project, the appetite comes while you´re eating".

VDMA Software und Digitalisierung also has answers for companies interested in machine learning. They are collected in an "Industry Podcast". In the first episode Peter Seeberg of Softing Industrial explains how a company should start a machine learning project. He also explains which know-how is required and which infrastructure is needed.

Basically, machine learning can be used to optimize product characteristics as well as internal processes. The characteristics of machine learning also differ with the products: on the one hand, these are located in the product itself, and on the other hand in the process environment of the machine, for example in the form of maintenance or additional value-added services.

One field of application of Machine Learning is machine operation. This is simplified by expert systems. A successful example of this is intelliGuide from HOMAG Plattenaufteiltechnik. IntelliGuide is an operator assistance system that reacts intelligently to the operator's actions. The system guides the operator using optical signals that appear directly in the work field of view. Within the full version, the machine operator benefits from all-round visual support from the intelligent system.

As a result, the machine builder's familiarization time, training effort and set-up times are reduced, while the machine operator's efficiency is increased at the same time. Machine learning thus enables both: the machine builder and his customers to optimize processes.

A topic for the future
Machine Learning enables technical systems to learn from experience. Algorithms are used for the system to recognize patterns and structures with example data provided by humans. Machine Learning then applies this new knowledge to new, unknown cases.

The VDMA Software and Digitization helps companies to successfully shape the path of machine learning with VDMA members. In its network, the professional association has a large number of companies that already have technological knowledge about machine learning. This knowledge should be used profitably for machine and plant construction. The Machine Learning Expert Group has been working on publications and assistance for VDMA members for three years now. New use cases are developed in regular meetings. The current publication is the Quick Guide Machine Learning.

Quick Guide Machine Learning

With the aim of helping all members with the business assessment and relevance of machine learning in order to enable their own approach and strategy definition, the professional association has published the Quick Guide "Machine Learning". The Quick guide helps to view opportunities, benefits and risks of machine learning in a structured way.
adding all to cart
File added to media cart.