Protecting your data is a top priority for ai-omatic. We are currently in the ISO 27001 certification process and use proven technologies to ensure the security of your data. These include TLS encryption, use of the Azure Cloud and IP whitelisting.

ai-omatics technology differs significantly from the solutions of competing companies in terms of its scalability and the accuracy of the results. To do this, the digital maintenance assistant developed by ai-omatic uses the engineer's specialist knowledge of the causal dependency structure between machine variables and combines this domain knowledge with the power of machine learning. This allows for higher model accuracy and more explainable results, increasing the acceptance of the approach among engineers. The result: - 20% better prediction accuracy - 70% faster scalability - an explainable AI solution for predictive maintenance

The customer receives access to a web-based dashboard on which the health score of the machine is displayed in the form of a factor between 0% and 100%. Through this, ai-omatics customers get an insight into the condition of their machines and can trace abnormal machine behavior down to the sensor level. By means of a trend display, the dashboard also offers the user the option of evaluating the forecasts and assessing value groups.

Use of existing data: ai-omatic uses data that is already collected from the machine or the process. This means that no additional sensors are required. This significantly reduces costs and installation effort. Taking into account various data points: In contrast to pure vibration monitoring, the maintenance assistant can include data points that have already been recorded. This makes it possible to monitor machines where vibration patterns do not provide meaningful information or where components within complex machines are difficult to access. Inclusion of external influencing factors: The biggest difference is that the ai-omatic maintenance assistant includes external influencing factors, such as different operating modes or the current product recipe, in the analysis. Unlike sensors that are not synchronized with the process, ai-omatic can more closely monitor the "normal state" of the machine. This enables a more precise detection of changes and an early indication of possible problems. ai-omatic learns individual vibration patterns for each distinguishable operating mode, which ensures high sensitivity and accuracy.

ai-omatic is able to process different types of data as long as they are available via a suitable interface (especially OPC UA). This includes numbers, decimals, text, and true/false values. ai-omatics software takes into account both sensor data and the expertise of the engineer, which can be entered via a simple interface. By combining these two types of data it is possible to provide effective maintenance.

ai-omatic is able to monitor the behavior of machines and detect even the slightest deviations in the data. Taking the contextual factors into account, a risk assessment of the abnormal machine behavior can be made, alerting users to adverse machine behavior. This ensures, among other things, the following: - Early detection of machine or individual component failures - Identifying the causes of the erroneous behavior - Fast troubleshooting and avoidance of consequential damage and costs

The data can be accessed using two possible options. As a rule, the data is transmitted and stored using an OPC UA live data source. Alternatively, a specially designed IoT gateway is used, which takes into account various data sources such as vibration sensors, analog and digital signals and measured process values. The connection is made via mobile radio, the information security is ensured by means of a data diode.

ai-omatic's algorithm does not require any historical sensor data or error logs, but learns from new data that is sent to the system during live operation.

In order to use the ai-omatic software, certain technical hardware requirements are required. The customer should have an Ubuntu server that has access to the OPC UA server. In addition, the following technical specifications should be met: The following minimum hardware specifications are recommended for a VM with local storage: Operating system: Linux Ubuntu 20.04 Processor: 4 vCPU Memory: 16GB RAM Storage: 250GB SSD hard drive The following minimum hardware specifications are recommended for a VM without local storage: Operating system: Linux Ubuntu 20.04 Processor: 2 vCPU Memory: 8 GB RAM Storage: 70GB SSD hard drive The following network requirements should also be observed: Outgoing ports: 443 MQTT/AMQP via websockets, 5671 AMQP, 8883 MQTT The server should be in the same subnet as the OPC UA server Sufficient bandwidth must be guaranteed after consultation.

Identify a machine, plant or process that you want to monitor. Connect the machine or PLC with an OPC-UA interface and the appropriate credentials. Identify the paths of the sensors associated with step 1. Make sure you have an Ubuntu (virtual) machine that can connect to the OPC UA interface. For more information, see: https://learn.microsoft.com/en-us/azure/iot-edge/production-checklist?view=iotedge-1.4#allow-connections-from-iot-edge-devices Install the Microsoft IoT Edge Runtime and configure it according to the needs of your network. For more information, see: https://learn.microsoft.com/en-us/azure/iot-edge/how-to-provision-single-device-linux-symmetric?view=iotedge-1.4&tabs =azure-portal%2Cubuntu Also note any required proxy server settings, which are described here: https://learn.microsoft.com/en-us/azure/iot-edge/ how-to-configure-proxy-support?view=iotedge-1.4