Digitalisation

With recent advances in technology, the rail sector has accumulated a wealth of data about its operations, which offers opportunities for improvements across the networks.

Yet, making sense of the flood of digital data and associated complexities is not a straightforward task.

Rail asset performance specialist Andrew Smith said it is vital that the industry make the most of the latest innovations in data collection and analysis to help maximise rail efficiency.

And he should know. As the rail portfolio manager with infrastructure engineering software development company Bentley Systems, Smith has spent three decades working on rail operations and maintenance issues, specifically on linear decision support.

Challenges faced by railways

One of the key challenges facing the railway sector, is that the requirements on them are continuously changing and evolving.

“Owner-operators are getting more trains. They’re getting longer trains. They’re getting longer operating hours pushed on them as well,” Smith said. “And generally, the budgets aren’t increasing. If anything, they’re decreasing. So, there is less time and budget to do the required maintenance work on an asset that is deteriorating more rapidly because it is getting more intense use.”

Owner-operators cannot simply do more of what they were doing before with fewer resources. Instead, they have to work smarter by finding inefficiencies in existing processes, so that they can evolve the way they do their work.

“We need to move from a reactive world of fixing something after it is broken to a predictive world where we intervene before the asset fails,” Smith said.

What data can be used to support rail operations and maintenance?

“We already have a vast amount of data about what’s going on across our networks in terms of usage, maintenance, inspections, measurements and in many cases, its design. What we are not doing is taking full advantage of that data in order to be able to make optimal decisions,” Smith said.

“Understanding the data is important. We’ve got all these disparate data sources and really need to understand the quality of them first – the limitations and the value – and make sure that we’re only providing end users with information derived from data in which we have confidence.”

How can technology help rail engineers overcome the multiple challenges of maintaining a safe, reliable network?

Bentley supports the professional needs of those responsible for creating and managing the world’s infrastructure, including the railway sector. The company delivers solutions for the entire infrastructure asset lifecycle, tailored to the needs of the various professions who will work on, and work with those assets over their lifetime, such as the engineers, planners, contractors, managers, and operators.

“We provide engineers with capabilities through intelligent linear analytics that enable them to work differently to increase efficiencies and reliability. However, the use of technology has to go hand in hand with process changes to realize the full value from an organisation’s digital investment,” Smith said.

Comprised of integrated applications and services built on an open platform, each solution Bentley develops is designed to ensure that information flows between project team members to enable interoperability and collaboration. Railways have an abundance of data about their asset, but it is hidden in spreadsheets and databases, and pieces of paper all over the place – so it is fragmented.

Even simple examples such as the ability to see the current condition of the asset, plus the work associated with it at the same time, have not been practical. These disparate datasets need to be brought together into a single-source solution. That involves taking data from measurement systems that come in from inspections, work orders, asset registers and a myriad of other classes of data. Then that data is federated to improve the rail operator’s decision-making.

“Engineers don’t want data: it’s just numbers,” Smith said. “They want the data translated into meaningful information. Such as what’s the history of what’s happened? What’s the environment? What’s the data actually saying about the real world?”

When the data is federated in a single solution, additional information can be overlaid – including relationships between the different data sets and formats, to provide the additional insight needed to deliver improved business outcomes.

Example in action

Recently, Bentley announced that SMRT Trains, the pioneer Mass Rapid Transit (MRT) operator in Singapore that operates over 282 kilometres of rail track, successfully completed the implementation of a Predictive Decision Support System (PDSS) for Singapore’s North-South and East-West lines, the oldest MRT lines in the country.

The partnership between the two companies will allow for a rail predictive maintenance solution that visualises all rail asset information and manages, monitors, and analyses rail conditions to help keep the tracks in good condition and avoid delays.

As of August 2019, SMRT Trains achieved its 1 million mean kilometres between failure (MKBF) target. Data correlation is twice as fast and with easy access to data they have significantly streamlined multiple analyses. With prioritization implementation, SMRT cut hundreds of manual planning hours and saved about 20 maintenance train deployments annually. Engineers also have a better idea of work conditions, allowing them to improve preparation and save time on the site.

The end game

Collaboration with the end users is key for the system to achieve maximum benefits.

“The first thing we typically do is review the way the rail operator currently works and then encapsulate that within the solution,” Smith said. “Then we work with the users to identify a series of incremental process changes they need to put in place to drive improvements.”

Once they are familiar with the solution itself, they can identify areas on the network where faults keep recurring, dive deep into those and assess how effectively they’re coping with them. The solution will actually help SMRT Trains to change the way they work, the way they look at data, and help them to assess how effective their work has been.

“Armed with this insight, many of our users have been through multiple steps of incremental, measurable process changes and improvements,” Smith said. “It can’t be done purely by technology. There has to be input from the engineers, and the priorities are always different from one user to the next, as they have different challenges to overcome.”

Operators are responsible for a large, safety critical system, not just a series of assets. The rail network needs to be treated as a system that is managed and maintained, and where one can make decisions that help to build and optimise what is being done underneath microscopically.