In-service monitoring should be the new norm

With increasingly complex demands, the railway industry faces significant challenges, including an ageing fleet and infrastructure, while needing to maintain operational efficiency and reliability. Traditional maintenance approaches, often reactive or schedule-based, struggle to meet these demands. To address these challenges, in-service monitoring using inertial sensors is emerging as a cost-effective, transformative approach, enabling predictive maintenance through a detailed understanding of vehicle and track interaction. This approach looks at the railway as a single system, involving both track and train, providing a holistic solution to ride quality and track geometry issues for early fault identification and proactive repair.

How in-service monitoring works

In-service monitoring transforms passenger and freight trains into continuous data collectors, capturing the dynamic interaction between track and train. Sensors, such as inertial measurement units, installed on vehicles, measure parameters like track alignment, vehicle stability and suspension performance. This enables early fault detection, predictive maintenance and supports root cause analysis.

Unlike periodic inspections or manual reviews, which may overlook early signs of wear or degradation, in-service monitoring continuously gathers real-time data from sensors embedded in trains. These data offer detailed insights into infrastructure and vehicle conditions, such as track alignment, vehicle response, and suspension dynamics. By identifying early signs of wear and correlating data between wheelsets, bogies, and the vehicle body, operators can take pre-emptive action to address issues before they escalate. A key advantage of this method is that it evaluates train dynamics during daily operations. Variations in wheel-rail interaction directly influence the dynamics and stability of the vehicle, helping identify areas where small irregularities might cause noticeable excitation of the vehicle. This information supports maintenance regimes by pinpointing locations and assets that require attention

Precision data for predictive maintenance

There are several approaches to analysing the data. Model-based and mathematical techniques can provide reliable, repeatable and comparable results that align with inspection vehicle standards, offering actionable insights. Alongside or alternatively, trend and feature analysis using machine learning and artificial intelligence (AI) can identify patterns that signal specific faults. In-service monitoring offers a predictive approach, using precise data to forecast faults before they cause significant disruptions. This enables operators to optimise maintenance schedules, minimise emergency repairs, and enhance the management of rolling stock and infrastructure.

Impacts on passenger comfort and operational efficiency

In-service monitoring delivers benefits for passengers primarily by reducing service disruptions. Rail operators can use this technology to address potential issues before they affect travel, ensuring smoother and more dependable services. Infrastructure managers and train operators currently conduct cab rides to analyse ride quality and identify track or vehicle issues, requiring trained personnel to be physically present. In-service monitoring automates this process, which saves operators both time and effort through accurate fault localisation. Additionally, by identifying and addressing sources of excessive excitation, in-service monitoring results in a more comfortable ride for passengers.

The data can assist in decision-making regarding rough ride reports and speed restrictions. Some reports from train drivers may be caused by poor wheel profiles, which do not require speed restrictions for the line. In such cases, the train operator needs to address the vehicle-related issue. Where track degradation is the cause, the data can pinpoint the location and type of fault. This also assists with the current regime and the unavailability of the measurement fleet, such as the NMT. It can fill the gap between NMT runs, help with better degradation modelling and distinguish the importance of measuring track conditions from the perspective of the vehicles running on it daily.

Looking to the future: setting a new norm in railway maintenance

The adoption of in-service monitoring represents a logical shift in railway maintenance, establishing a forward-thinking, data-driven approach that manages rail networks as interconnected systems. Treating the railway as a cohesive system involving both track and train ensures that maintenance activities address the root causes of ride quality and track geometry issues comprehensively, leading to a more resilient and sustainable railway system. By transforming trains into intelligent, in-service data collectors, the rail industry can take a significant step towards modern, reliable and resource-efficient maintenance practices.