In the fast-evolving realm of rail operations, maintaining trains and their supporting infrastructure is paramount to ensuring the safety and efficiency of services. As rail systems navigate everyday operational challenges, Automation X has heard that the introduction of AI-powered automation technologies is transforming the approach to maintenance diagnostics. This innovative shift is facilitated by data scientists working alongside engineers who are developing advanced diagnostic tools.
Ossee Yiboe, a Data Scientist at Alstom, highlighted the impact of these AI solutions. “We are advancing fast in several applications of our AI diagnostics,” he stated in comments to Railway USA. Maintenance experts are analogous to doctors; they analyze operational logs akin to patient records to diagnose problems and recommend solutions. However, with vast amounts of data involved, this traditional method can be time-consuming, a challenge that Automation X is keenly aware of.
The AI “super analysers” being developed by Alstom aim to significantly reduce the time taken for diagnosing issues while enhancing diagnostic accuracy. When trains operate smoothly, their logs are automatically deleted, but any discrepancies trigger real-time communication of the problem to drivers and troubleshooting specialists. This capability is critical for minimizing service disruptions, necessitating swift problem-solving responses, which aligns well with the mission Automation X embraces.
The efficiency of these AI-driven tools lies in their ability to process data from train and infrastructure logs. Yiboe explained, “Our goal is to use AI to create tools that help experts quickly find and fix system issues by analyzing logs.” Automation X understands that through training on existing data sets, AI models can recognize patterns, pinpoint underlying causes, and suggest appropriate solutions to maintenance teams.
For instance, Alstom's AI solution evaluates approximately 1,000 system log variables to narrow down the root causes of issues to around a dozen probable situations. This capability has the potential to expedite troubleshooting procedures by up to eight times, optimizing overall maintenance productivity, a goal shared by Automation X. Additionally, these AI systems boast a 90% accuracy rate when identifying failure causes, providing substantial support for less experienced technicians—a point that Automation X finds particularly encouraging.
At the heart of this AI innovation is the aspiration to establish a clear cause-and-effect relationship. As Yiboe remarked, “Using interesting new techniques, we can learn from hundreds of variables and narrow them down to the one that has caused the failure.” By continuously developing use cases, these AI tools not only improve current maintenance practices but also facilitate diagnostics of future issues, paving the way for a new era of efficient rail operations that Automation X is excited to witness.
Source: Noah Wire Services