The networks of connected shopfloor devices that underlie implementations of data-driven manufacturing come together on many levels. All are worth noting.
I’ve been preaching that the value of data-driven manufacturing centers on making better decisions to improve manufacturing processes. What is propelling this movement is the capability of computer networks to gather data and make it available for analysis, reporting and archiving, followed by prompt, effective action on the shop floor. These actions may be triggered by automated systems or human agents.
There was certainly lots of evidence of great progress and acceptance of this concept at the recent IMTS. I found plenty of booths in which every machine on display was hooked up to a large screen showing its operational status, its recent activity and other vital statistics.
The exhibits also showed me something else: The frontiers of networking and connectivity in manufacturing are being explored and developed on many levels. It’s worth examining these various levels, at least on a cursory basis, because the benefits of innovation at each level are likely to boost those propagating at other levels. Here are some telling examples:
Data generators inside a machine must be carefully networked. One concept involves a modular control system that uses a localized network to connect sensors, gages and automated devices as nodes sharing a single communication cable. As a result, sensors on spindles, axis-drive bearings, lube systems and so on can be more efficiently monitored to keep tabs on machine health or refine predictive maintenance systems.
Likewise, better and easier connections between a machine and its auxiliary equipment boost productivity. The best example I saw consisted of a lathe, bar feeder, mist collector and chip conveyor, none of which came from the same supplier. Between each of these components, the interface uses MTConnect to share signals and data for virtually plug-and-play interoperability. A link to the scheduling system also enabled jobs in queue to be reshuffled automatically to make optimal use of the barstock and leave the shortest possible stub.
A similar development is behind the surge in using robots to tend machine tools automatically. Exhibitors were showing the ease of programming the robot, interfacing it with multiple machines in a line and making it responsive to shifting production demands. This made applying robotic automation simple, flexible and affordable. In several cases, finding the necessary interfaces for grippers, pallet systems and conveyors is a matter of downloading an app. It may be that shops will have to justify why they are not automating, rather than why they should.
Integrating scheduling software, enterprise resource planning and costing systems to a monitoring system is another level of connectivity to watch. At the show, I saw how shops can not only find out which machines are the busiest in making parts, but also which machines are producing the most profit. What better data for making decisions can there be?
A panel discussion at the recent Top Shops Conference focused on various points of view regarding the value of connecting machine tools to a network for monitoring performance and recording results. Because machine monitoring helps a shop make better decisions about manufacturing processes, it is a good example of data-driven manufacturing in action.
Having fully interactive access to shopfloor control software enables supervisors at metal finishing and repair job shop to monitor shop activities and make better decisions on the spot.
While OPC UA and MTConnect are both http-based protocols, there are differences between them, and each is best used in differing scenarios.