Searching for a ‘data graveyards’ solution
Railways have yet to fully realise the benefits of making best use of operational information collected – much of which is still ending up in ‘data graveyards’. That will be one of the main messages of railway data analytics expert Pietro Pace at this week’s Intelligent Rail Summit.
From his position with MERMEC, one of the leading providers of rail diagnostics and inspection services, Pace asserts that condition monitoring – essentially the process of determining the condition of infrastructure and rolling stock – can provide crucial information that enables operators to optimise their management and maintenance of their assets.
It was, and in many cases still is, the case that infrastructure data was being collected, checked against the appropriate safety parameters and that was it. It was only used once, and thereafter left in the ‘data graveyard’, never to be seen again.
That, says Pace, is where condition monitoring can make a critical difference. Rather than just viewing data sets in isolation, condition monitoring measures against several parameters to identify any significant changes in assets or components that may, left unchecked, severely affect operations. Operators can take data from multiple conditioning devices and, through rigorous analysis, identify any safety-critical defects and recommend appropriate maintenance and renewal works.
“MERMEC’s use of multiple condition monitoring devices includes performing status checks on track geometry, rail profile and clearance; patrolling the track for evidence of missing or broken components; and continual testing of the interaction between train and track,” says Pace.
“The constantly growing demand for more efficient and cost-effective management of railways should be putting pressure on operators and maintainers to make the best possible use of their asset data,” adds Pace, who is Product Line Manager in MERMEC’s Advanced Services Business Unit. He is among a host of expert speakers presenting at the summit in Naples, from November 22-24.
“Recently I have been spending more and more time with railways in all sectors – passenger, metro, freight and heavy haul – dealing with the need to manage large amounts of data from multiple condition monitoring devices. This has been to identify objects and defects, and estimate the health and condition of assets, in order to recommend the required inspection, maintenance and renewal work.
“Other devices can be installed on the track to monitor the trains. For example, measuring and inspecting the wheel profile and surface, brake pads, shoe gear (structure including contact pad used by third rail power supply), train integrity before a tunnel (motivated by the need to stop rolling stock in time in case of a fault), axle box temperature and pantograph. From optoelectronic (sourcing, detecting and controlling light) systems used for the accurate measurement and high-speed inspection of the track’s parameters, to image processing algorithms fused with laser technology, today we can cover every possible aspect of the overhead line infrastructure, as well as the health of the rolling stock components.”
As to whether companies – and in particular rail freight operators – are fully seeing the potential long-term economic benefits, Pace is keen to emphasise that there is so much more that could be, and in MERMEC’s case, already is being done.
“We can observe some differences of breadth and maturity in the way big data is being used for asset management. Some railways are indeed making the best use of their data, whereas others are only beginning to explore the possibilities,” he says. “We have sold more than 1,500 measurement and inspection systems worldwide, and see ourselves as contributing significantly to the realisation of such economic benefits.”
A long-time proponent of tackling ‘data graveyards’, Pace adds: “This is a topic I have been talking about for many years. The need for efficient and effective use of data to drive infrastructure and rolling stock maintenance has long been recognised. There is an increased number of conditioning monitoring devices producing big data.
“Turning this large amount of data into value is what ‘digitalisation’ is all about, for the railways as much as for all the other industries. From the complete automated digitalisation of all the rail assets to new paradigms of rail infrastructure diagnostics, based on the ‘Internet of Things’ approach – a worldwide vision of an internet in which everyday objects have network connectivity, allowing them to send and receive data – digitalisation shows a tremendous potential towards optimisation.
MERMEC duly named its new integrated model for infrastructure diagnostics the ‘Internet of Tracks’, or IOT. Significant advantages are ‘real time’ information on the quality of the infrastructure; fewer disruptions in regular train traffic and operations; and more efficient use of people for track inspections.
Pace also asserts that any barriers to successful implementation go beyond the introduction of condition monitoring devices: “The key successful factors for implementation include the level of interaction between hardware and software, and the processes and strategies adopted by rail operators for extracting information from data. Based on my own experiences, these make a great difference between dealing with “big data” and “big” information.
“To take full advantage of data from multiple devices means they must be integrated so that data can be visualised simultaneously to ensure immediate correlation. Priority to the user experience is crucial to ensure the highest performance in all critical phases: acquisition, processing, accessing, viewing and reporting.”
Citing a famous quote by computing pioneer Alan Kay, he flips it around to suggest the precise opposite: “I would say that people who are really serious about hardware should make their own software, and MERMEC has introduced new software to deal with the variety of data collected by the devices. Traditional software was a barrier because it was operating in isolation.”
Ultimately, says Pace, it is all about enabling the railway sector to do things better and not create unnecessary additional work: “Any hardware device or data analytics software that is introduced in a railway organisation has to be considered not as a replacement of the human job but as a tool to allow operators to work smarter, and not harder.
“Several manual tasks such as patrolling track or manually correlating data with spreadsheets can be automated by systems which free operators’ time to let them focus on the new information extracted that, ultimately, creates more value for their organisation.”
Pietro Pace’s presentation will be on day three (entitled Big Data in Railway Operations) of the Intelligent Rail Summit.