(Image: ©Dirk Ercken)
The seamless networking of data, devices and systems is an essential prerequisite for Industry 4.0. Ideally, the various components (sensors, devices, systems, etc.) should not just collect data locally, but make it available to other components as a service. In purely technical terms, we are rapidly approaching this vision. For example, we can seamlessly track the path of a component from receiving to the serial number of the product in which it was installed. We can also log process parameters like soldering profiles or the force used to place the component on the board.
Does that mean that everything is okay? I don’t think so, because most of what we have today are partial and proprietary solutions, which means that all the components in a smart factory must come from a single manufacturer and his partners.
To break through these borders, we need new ways of cooperating with each other.
The makers of SMT manufacturing equipment, i.e. the producers of reflow ovens, AOI systems, placement machines, printers, material management solutions, etc. have extensive process expertise. They know what data is relevant for churning out top-quality products and how to collect and use it.
The first hurdle is the fact that many equipment makers don’t make the data which their machines collect available via open interfaces. An open interface in this context means that all process data collected by the machine is published as a service and can be accessed by any other manufacturer and/or device. I don’t know many equipment makers who are able (and wanting) to do this today.
The second hurdle, in my opinion, is the fact that many machine manufacturers lack the know-how needed to network and process the vast and unstructured amounts of data being generated.
That’s why I urge my fellow software developers in the machine engineering field to build alliances with classic IT firms. The sooner we do this, the better, because these firms have much more expertise in the areas of database management, the handling of unstructured data, big data, etc. And I am confident that our colleagues in the IT industry are interested in this as well. Especially in the United States, IT forms view industrial applications as a growth opportunity. They talk about the industrial internet, but as they rush into these applications, they often forget that they lack the specific process know-how and expertise with regard to the special requirements of real-time data processing in industrial controls. This is where the IT industry in turn needs our knowledge of processes and controllers, as well as our experience in interpreting the data being generated in SMTproduction.
I am convinced that significantly enhanced cooperation between industrial engineers and the IT industry is needed to implement the smart SMT factory.
For an interesting overview on software, interfaces, etc., you may also want to read the following article on a panel discussion at www.all-electronics.de.