Machine breakdowns are a common bottleneck in factories to reach optimal production efficiencies and achieve tighter production deadlines. Although machine downtime cannot be completely avoided, critical breakdowns and longer downtimes can be avoided through the clever use of IoT sensory data that is used along with state-of-the-art artificial intelligence-based anomaly detection algorithms.
Xeptagon in line with the latest advances of Industry 4.0 is launching a core software module on Predictive Maintenance for factory machines. The software module is integrated with the Xeptagon IPS system that is developed for intelligent Capacity Planning and Production Scheduling of factory floors. The integration will enhance the capabilities of the IPS system by further improving factory floor production efficiencies while reducing costs.
The Xeptagon Predictive Maintenance system includes support for machine Condition Monitoring as well. The system includes an easy-to-use dashboard that will enable maintenance engineers to remotely monitor machine vibrations, temperature, humidity, and other machine operations. The dashboard will immediately indicate any abnormalities of these observations as alarms where the maintenance engineers can quickly attend to the machines to avoid any critical breakdowns reducing the machine downtimes. In addition, the system allows customisable alarm thresholds for the above parameters as required by the maintenance engineers.
The core part of the Xeptagon Predictive Maintenance component is the capability to further analyse the IoT data captured through AI-based algorithms to evaluate the working conditions of the machines. These algorithms allow early identify machine faults, categorize faults, and even estimate the time to the next machine failure is likely to occur. As a result, maintenance engineers can immediately attend to these machines at an early stage of the fault before the actual breakdown happens. The maintenance engineers can take short-term remedial actions as well as prepare spare parts and repair teams to attend to the issue before production comes to a complete halt. This will help factories to avoid and reduce downtime to a minimum level while reducing maintenance costs.
Since the Xeptagon Predictive Maintenance component is integrated with the IPS system, project managers in factories are immediately notified of machine breakdowns and alterations of the production schedules as a result of the machine breakdown. The system will intelligently reroute production schedules to the best possible scenarios in a manner that the most critical production deadlines can be achieved.
Xeptagon Predictive Maintenance utilizes the latest Plug and Play IoT devices that can be remotely configured and installed by maintenance engineers with minimal training. Our IoT devices come with a manufacturer's warranty as well as a performance guarantee to suit harsh factory environments. Xeptagon utilizes state of the are vibrational sensors using LoRaWAN communication protocol for long-range communication and supports the concurrent monitoring of hundreds of machines. In addition, the IoT sensors are connected to the AWS IoT Core of Amazon Web Services to get the advantage of the latest cloud technologies.
The Xeptagon Predictive Maintenance also includes an Active Learning-based AI component which enables the system to continuously learn from human input. For example, if the system falsely identifies a fault that doesn’t exist or misses to identify an actual fault on time, the maintenance engineers can provide feedback to the system on these issues. In such scenarios, the system will incorporate the feedback to further improve prediction capabilities in the future.
The system is currently tested in the manufacturing plants of Sierra Cable PLC, a listed company on the Colombo Stock Exchange (CSE) with production facilities in Sri Lanka, Kenya and Fiji.
Read more details about the Xeptagon Intelligent Production Scheduling (IPS) System here.
Need to Predictive Maintenance Systems for your Machinery?
Get in touch with us!