Germany-based Software AG has released a new self-service analytics tool to enable operational staff to rapidly spin-up analysis of real-time streaming data from industrial machines and solve and predict production issues.
The new solution, called Analytics Builder, is effectively a skin for Software AG’s Apama streaming analytics product, which packages-up its capabilities as a set of modular ‘building blocks’ and extends its use to the shop floor, and beyond the realm of programmers and data scientists.
Software AG described the solution as a “game-changer” for streaming analytics, a centrepiece of the standard industrial IoT template for predictive maintenance and production-line scheduling. Analytics Builder will reduce the time it takes for industrial companies to design, develop, and deploy analytics for IoT applications, it said.
Jeremy Hill, director of product marketing at Software AG, commented: “Rather than have programmers write code to take advantage of that streamed data, engineers can build the analytics themsleves. It is a game-changer. What you can do with it is pretty much unlimited.”
Stefan Sigg, chief product officer of Software AG, said: “By allowing operators and domain experts to directly develop their own analytic applications we are taking a huge amount of pressure off the IT department and empowering de-centralised teams to accelerate their delivery schedules.”
Analytics Builder features a graphical user interface for non-coders to take advantage of streaming analytics. It allows operators to quickly build and deploy analytical models using a library of pre-built analytics blocks, a drag-and-drop web-based editor, and the capability to manage deployment and simulation.
The new solution is available via the Cumulocity IoT platform, which is available on the IoT outlets of Alibaba, AWS, and Azure, and via reseller channels, including distributor Tech Data, and various telecoms operators, which offer Cumulocity as a white-label IoT platform.
It may yet become available via third-party IoT platforms, as well. “It is someting we are looking at seriously. It will come down to priorities when we come to it, but it is something we are looking at,” said Hill.
Self-service IoT is key plank in Software AG’s business transformation, as it seeks to popularise its software tools for integrating “data, devices, clouds, and applications” for enterprises.
Its Cumulocity IoT platform makes available device and connectivity management and data analytics as self-service tools. Its TrendMiner application, for identifying patterns in historical data, also breaks out its offer as self-service tools. “You can connect a device, update its firmware, feed into our analytics, and pass the information to your enterprise systems in about 20 mins, if want,” said Hill.
This self-service approach puts IT tools into the hands of the OT brigade, with hard-earned domanin expertise, to direct their own digital-industrial detective work from the shop floor. “It makes it possible for the operation managers, who understand the production line, to find out what has gone wrong, and what could go wrong, and try to fix things.”
Apama Analytics Builder is a product of Software AG’s collaboration with the ADAMOS group, a venture with Germany-based machine building companies DMG MORI, Dürr, and ZEISS, which focuses on the joint development of software-based industrial IoT solutions.
The requirement for a simpler analytics model was identified by Dürr, which makes painting robots for the car industry, and other manufacturing sectors.
Nico Koch, manager digital factory at Dürr, said: “[This] will push forward the digitalisation on the shop floor tremendously. We have put extremely powerful analytical tools into the hands of those that understand the production process the best, opening a new era of streamlining and improvement”.
Hill commented: “Dürr wanted its engineers to be able to create analytcs applications, without having to programme them, in order to spot defects in paintwork and painting processes during production – instead of during inspection afterwards.”
Dürr’s customers, using its painting robots, can now stop the production line when a defect is detected, rather than re-running the process, and respraying every car.
But the dialogue with Dürr was just the start point. The blocks in Analytics Builder habe developed, and its application stretches beyond painting robots to capture a variety of standard industrial processes. “It has expanded, and caters to all kinds of examples.” It will develop further, as new requirements emerge. “But it is a robust set, as it stands; there is an awful lot in there, now,” addede Hill.
Hill has a tale about the high cost of undetected errors in production, which is anecdotal, but worth retelling again. “There was a great example, I heard, about an oil refinery – that it costs $10-$20 million per day for a planned shut-down, and nearer $10-$20 million per hour for an unplanned shut-down.”
He added: “The point is the ability to do analytics on real-time data is invaluable; if you can spot an event before it happens, then the saving is massice. The reason streaming analytics hasn’t seen the momentum up to now is because it has been too complex. We are trying to make the complexity go away.”
The idea idnustrial IoT analytics is commoditised with the emergence of modular self-service tools like Analytics Builder should be welcomed, suggested Hil.
“Data scientists will be more focused on where they add value than on the bits that can be done in Analytics Builder. If that means a lot of analytics become commodities, then that is a psoitive. If the value in the data is critical for people tio get value out of IoT strategies, then if we can do that more quickly for organisations, then that can only be a good thing.”
