Four steps to creating your own cloud app
Industry 4.0 supplies enormous quantities of sensor data. Cloud technologies and cloud apps for analyzing this data are becoming the key to greater efficiency and productivity. Here’s a real-life example from the Siemens electronics factory in Amberg, Germany.
Are you familiar with Amberg? Situated among gently rolling hills, the city of 42,000 residents blends into the landscape halfway between Nuremberg in Southern Germany and the Czech Republic. It’s no Silicon Valley for the high-tech industry, or so it would appear. But the first impression is deceptive. Actually, this is the place to come if you want to glimpse the future.
Siemens has been producing its famous Simatic controllers in the Amberg electronics factory since 1990. The plant manufactures 1,200 different types, and 120 variants roll off the assembly line every day. One product is finished every second. The 3-shift facility produces a total of around 17 million Simatic products each year. The Simatic controller is the brain of a machine or production line. The manufacturing facility for this controller in Amberg represents the digital factory of the future, where the real and virtual production worlds merge to achieve one primary goal: to deliver fault-free products.
50 million data records per day
To achieve this goal, everything in Amberg is networked with everything else, and the entire production process is translated into data. All components can be identified, and each product is given its own unique number. Sensors document all steps throughout the manufacturing process in real time and collect product information such as soldering temperature, device placement data, and test results. As a result, more than 50 million data records accumulate every day.
Connecting to the Internet of Things
This is an enormous wealth of data just waiting to be used. Connecting the machines to the Internet of Things (IoT) plays a central role here. In fact, due to the increasingly complex production structures with their enormous data pools, existing analysis methods are now reaching their limits. New technologies, such as cloud platforms, edge computing, and artificial intelligence, are needed to collect a vast amount of data and analyze these complex interrelationships quickly and systematically.
Quality, performance, and availability at a glance
With the aid of MindSphere, the Siemens cloud platform, a project team in Amberg has now developed an app that visualizes the productivity of a plant at a glance. It measures the quality, performance, and availability of an assembly line. One of the benefits of the new, cloud-based solution is that the application can be implemented easily and quickly, and the data points on all machinery can be connected. This facilitates the standardized calculation of the overall equipment effectiveness (OEE) from the collected raw data in a way that prevents manipulation.
Although the previous solution could also carry out complex analyses, it was expensive and complicated to implement. A second disadvantage was that it could not be upgraded on all equipment across the board, due to technical limitations. And it was a purely local solutions, which means the user could access the information only via the factory intranet in Amberg.
MindSphere and the new MindSphere app now make it possible to easily combine and visualize the relevant data from the sensors on the machines in Amberg. This information can be downloaded to a smartphone, PC, or tablet at any time and independently of location, even outside the plant for example. This enables factory-comperhensive performance comparison at a glance.
On this basis, the responsible teams can answer questions such as the following every day during the shop floor meeting:
- Is the equipment operating without downtimes? (Availability)
- How high is the output? Have we reached the set productivity target? If not, why not? (Performance)
- Is the end product okay? (Quality)
How the cloud app works
The cloud app delivers its results by following four main steps:
- Collect data and signals (IO) with a SIMATIC IOT2040
- Pre-processing of data with Node-RED
- Send 4 event-triggered values: Start trigger (bool), Stop trigger (bool), Article number and cycle time (int), Quality trigger (bool)
- Calculate tamper-proof KPI in the cloud (OEE = Availability x Performance x Quality)
Node-RED, a flow-based specific programming tool that runs on a Simatic IOT2040 control module or an edge device, is used to extract the data from the process. The devices collect and pre-process the data and then send the four data points to the cloud. The analysis in the cloud then makes it possible to compare different systems as well as departments and machines.
The transparency provided by the app can also be employed for predictive maintenance, thus permitting vulnerabilities to be identified and unscheduled downtimes to be prevented.
The MindSphere cloud platform is designed as an open operating system. This means that Siemens provides the platform, while the customers – in this case, Siemens’ own factory in Amberg – use existing MindSpere apps or develop their own applications as needed. They can then market new applications through a shared App Store.
Three MindSphere apps are currently in use in Amberg. In addition to the OEE app, developed in-house, for calculating and monitoring equipment performance, the newly developed Performance Insight app from the official Siemens App Store is also used for predictive maintenance. Performance Insight helps determine when a milling spindle (used to separate circuit boards) needs to be replaced. If a replacement is needed, the maintenance team is alerted via the cloud to prevent an unscheduled down time.
Collecting a wealth of data from production
Cloud technologies and cloud apps are thus becoming a central element of production control. They make it possible to use the extensive data pools that arise through the increasing use of digital solutions along the entire industrial production process.
It took the project team approximately six months to develop, test, and activate the OEE cloud app. However, they did this on the side and did not work on the project full time. The team first developed the app in MindSphere Developer Tenent and created a very first executable version. Then they connected it step by step in their Operator Tenant, to provide the App to the user community. Finally the App will be used operational in the user’s Value Plan.
So-called low-code platforms like Mendix are now used to speed up the development of apps for MindSphere. These platforms and the associated tools and services allow apps to be created more easily and up to ten times faster. This means that users no longer need to have explicit knowledge of programming language in order to develop an app.
We are seeing many projects being driven primarily by a combination of cloud technologies with edge computing and artificial intelligence.
So will everything be in the cloud in the future? Not exactly. In Amberg, we are seeing many projects being driven primarily by a combination of cloud technologies with edge computing and artificial intelligence. Industrial Edge facilitates electronic data processing and thus ideally adds processes such as MindSphere to cloud computing. Users can thus analyze the data more or less in real time on the field level right on the machine (edge computing) or send the information to the cloud for long-term analyses or app development, depending on their needs.
The first AI applications are also already in use in Amberg. AI based on edge devices helps reduce unscheduled machine down times due to motor bearing damage to circuit board milling machines by 100 percent, using predictive maintenance measures. The team is also working on reducing the amount of work involved in expensive X-ray tests for the circuit boards by 30 percent, using Simatic Edge and AI.
IT and OT (operational technology) are thus merging more and more with entirely new operator and business models. The Amberg team is already at work on the next cloud app. This time the app will be a fault analysis tool that helps operators improve the equipment.
Check out my blog to keep up to date!