Analytics | Predge Conveyor™️

Predge Rolling Stock™ Analytics features processes incoming data continuously to add further value to customers data generation. Hybrid models is the strategy for wear and fatigue driven failures with the aim of predicting events before they occur. An extensive understanding of root causes and a strong focus on validation of indications against documented events results in high hit rate with low number of false positives for our analytics.

The Wheel Damage Prediction (WDP) feature is based on AI and Machine learning principles with hybrid models. The WDP utilizes information from wayside detectors measuring wheel impact forces. By synchronizing it with estimated and measured distance, weather data, and tuning the performance against the maintenance records, a full-scale picture can be presented to the users, such as operation centers and maintenance planners.

The wheel wear prediction solution, WPP, replicates every wheel as a digital twin tracking the condition of the individual wheel, self-correcting itself and performing analytics as new data comes in but also predicts what next measurement should be expected from the field. The feature manages measurements provided handheld devices, one or many wayside devices or a combination of both. It can synchronize the data with distance information to gain results in distance rather than time when needed. Since it treats every wheelset as an individual it will account for different vehicle dynamics and contextual differences.

Load distribution on wagons might shift due to a combination of vibrations, lateral and longitudinal forces during operation. This phenomena can lead to negative consequences and in worst case cause derailments. By combining data from multiple detectors measuring axle load, changes in weight distribution in both directions can be identified. The ORLOS feature can provide an early indication on occuring load shifting but also introduce further information around how the load shifts on specific routes for individual wagons and complete train sets.

Understanding the data quality is a necessity when utilizing information as decision support. DPI provides an quality measure for incoming data sources and performance indications to compare sources, understand seasonal variations and detect deviations.