Intelligent asset management: Railways forging the future with cognitive computing

Business and Technology Writer

Intelligent asset management is a space where travel and transportation innovators are pushing the envelope, and no sector is more involved in data-driven transformation than the railway industry. Rail networks lend themselves to innovation: They are simultaneously dynamic and expanding while being grounded in fixed infrastructures that can be measured, analyzed and improved.

Cognitive computing is a powerful tool for rail leaders. From maintaining and developing whole systems to replacing single wheels on single train cars, the following examples illustrate ways that this technology is unfolding on a global scale.

Tight, tight, loose: Identifying track degradation

Infrastructure giants such as Network Rail, which operates most of the tracks crisscrossing England, know that patrolling kilometers of rail tracks requires significant man power. Using lasers and powerful computers, they're eliminating that drain by imaging moving trains down to 0.8 millimeter distances.

As Rail Engineer reports, the algorithms at work identify issues ranging from loose bolts to changing car conditions and even sub-optimal track geometry. Network Rail's human resources are than able focus on repairs and reformulations, rather than spending valuable time searching for problems in the first place.

In a recent International Railway Journal webinar, Keith Dierkx, global transportation and rail leader at IBM, described how video and print pattern-recognition capabilities can also detect wear and changes to wheels, trucks and rail cars. Further, analytics technology for travel and transportation providers can parse significant evidence of asset conditions from repair logs and maintenance e-mail correspondence, thereby streamlining operations and generating insights about what is happening currently and what is likely to happen in the future.

Rethinking the wheel: The train wheel, that is

In the United States, there are millions of individual train wheels that rail systems must monitor for faults on a daily basis. It turns out that watching wheels for failures is actually an acoustic data analytics equation:

  • Acoustic sensors coupled with computers can search for sound signals that suggest a damaged wheel. This means processing some 100,000 sensor-driven data points each day.
  • When the system finds anomalies, it triggers teams that locate the car in question and conduct maintenance before the crack ever reaches a critical point.
  • Bjorn Austraat, practice leader of North America for IBM Watson, points out in the International Railway webinar that maintenance crews can also get real-language repair recommendations from advanced analytics software, and they're able to receive and log new insights from the rail yard itself.

The process of monitoring wheels, which formerly relied on visual inspection, is undergoing a massive change, with data analytics altering the way rail companies perform intelligent asset management.

From alternators to wing rails: Assessing whole-life costs of rail assets

The London Underground (LU) will serve around 10 million people by 2030, according to Rail Technology Magazine. Numbers like that mean executives spend a lot of time thinking about the capacity of fleets. But, as LU's Head of Asset Strategy and Investment Andy Jinks recently told the publication, data analytics are allowing the agency to model whole-life costs against short-term and long-term expenses — sometimes reworking what it chooses to do in surprising ways.

"It may be that the best thing from a whole-life cost perspective is not practical or affordable or deliverable," he says. With the power of big data on their side, LU can assess the whole network and demonstrate how spending decisions impact all parts of it.

The future of intelligent asset management in railways

For now, especially in the world of freight distribution, intelligent asset management puts the focus on the tracks and not the cars. Sensors rely upon electrical power to function, and freight cars are typically not electrically equipped. As solar and other options gain popularity and complement sensor technology, however, that stands to change.

The development of even more data-driven infrastructure will almost certainly take root in the offices of rail CEOs as well. The cost benefits of plugging assets into cognitive-type data environments is well-evidenced by the vanguard innovators such as those we've just examined, who are already making strides in the railway sector.

Connect with analytics professionals to understand the full business potential of your data. Learn more on IBM's Travel and Transportation Industry Page.