Timely evaluation of tamping activities is crucial for infrastructure managers because not all activities result in the intended outcome. Sometimes, tamping results in an improvement that lasts only for a short period or no improvement at all. When ineffective tamping is left unnoticed until the next inspection, it can cause accelerated switch degradation or even a failure. Also, repeated unnecessary tampings can degrade switches further.
With the current approach of interval-based inspections, infrastructure managers have little opportunity to validate maintenance performance. To optimize maintenance practices, it is essential to continuously monitor the effect of maintenance and examine whether it was successful, for how long, and with which method.
In our endeavor for continuous improvement, we have gathered a wealth of customer feedback to fine-tune our maintenance validation feature to best support our customers’ needs. In this Fall ’21 release, we are re-introducing the Maintenance Quality Check (MQC) with enhanced accuracy, usability, and integration capability.
As one of the main features of the KONUX system, the MQC evaluates the tamping activities where the KONUX IIoT device is installed. With this feature, you can examine the tamping effectiveness in a matter of hours and plan an appropriate action in time in case additional measures are needed. Also, by analyzing the effects of different tamping methods, you can establish best practices.
The KONUX system continuously monitors the vertical displacement of the sleeper to measure the trackbed condition. The machine learning algorithms train the data models to assess recent maintenance activities against the past trackbed conditions or specific expectations defined by customers. The data models evaluate the tamping methods used and the condition before and after the activity to fully assess the executed maintenance quality.
If you have been following our releases, you might remember that in our Summer ‘21 release, we have presented the improved trackbed health model. As the insights from the MQC are drawn from the vertical displacement levels of the sleepers, the accuracy of the MQC chiefly relies on the trackbed health model.
As we integrate this advanced model into the MQC feature, it provides more accurate and stable insights on the effects of tamping activities, while filtering out irrelevant signals or factors that could cause false positives. As the detected signals are affected by various factors that are changing daily, such as temperature and train speed, the improved model reduces daily fluctuations of data points in order to provide stable insights.
The improved user interface (UI) helps you navigate the system easier and make decisions with all the context you need. The evaluation details for each activity provide all the necessary information to assess the maintenance performance fully – whether the tamping was effective, for how long, and the conditions before and after the action.
Now the absolute values are accompanied by the percentage values to enhance your understanding of the displacement improvement. Furthermore, the vertical displacement evolution since the tamping is performed is displayed in a graph together with the evaluation details. This helps you visually confirm the evolution of performance and understand better the context of the effectiveness over time.
Another element that adds to usability is the enriched tooltips and knowledge base contents that guide you through the system. As the KONUX system deals with highly complex and technical insights, easily accessible in-app guides can facilitate user experience throughout the system and provide useful and necessary details about the features or visual elements.
With today’s complex IT landscape, flexibility is key to maximizing customer value and achieving the long-term success of any system. As our customers deal with various systems to gather comprehensive data, integration is critical to help customers make decisions efficiently and act upon integrated insights.
The KONUX system supports flexible integration with other systems, allowing for enhanced usability in combination with other systems. One approach we support is widget-based integration with our customers’ asset management systems. Unlike sole integration of data points, widget integration enables widgets to appear seamlessly embedded within the customers’ systems, allowing integrated views of multiple systems in one place.
Widget integration allows for the following benefits:
Presenting insights in the most actionable way to help efficient decision-making (rather than simply feeding the aggregated data into the customer’s system)
Benefiting from tailored insights built on the best understanding of our models and data
Allowing users to directly benefit from our quarterly releases with new and improved features
Actively gathering ground truth from the users directly through the widgets to bring better functionality
The visibility into the effect of maintenance enables a holistic insight into trackbed health . With improved accuracy, usability, and integration capability, the MQC equips you with all the necessary insights to optimize your trackbed maintenance. By reducing costs spent on unnecessary activities, avoiding potential failures, and ultimately increasing asset lifetime, you can achieve enhanced availability and cost-efficiency.
Would you like to know more about these features in detail? Get in touch here.