15.06.15
Current IM fleet data collection service 'unsustainable'
Source: RTM Jun/Jul 15
David Goddard, head of Data Collection Service Enhancement at Network Rail, discusses the challenge of optimising infrastructure data measurement going forward.
Network Rail is starting to consider the feasibility of moving equipment from its infrastructure measurement (IM) fleet over to other forms of collection platforms, potentially including passenger and freight trains.
The current IM fleet comprises 13 trains formed of 64 carriages and operates over 2,000 shifts covering 18,000 miles of recording over 550,000 data-miles every year.
However, speaking at this year’s Railtex event, David Goddard, head of Data Collection Service Enhancement (DCSE) at Network Rail, said: “The fleet that we have got is quite old and with age comes reliability issues. We are seeing, quite increasingly, that the reliability is going down but the maintenance costs are going up.”
Network Rail’s IM fleet collects 18 different types of condition measurements (datastreams), using different technologies and collection modes to obtain the various measurements.
“Generally, though, they [the fleet] are not entirely fit for purpose,” he said. “We need to think about how we are going to do this for the future.”
Goddard highlighted that some of today’s legacy systems restrict asset condition data recording speed. For example, the existing rail flaw detection system uses ultrasonic technology, which limits the collection speed to 30mph.
The DCSE programme, part of the wider ORBIS (Offering Rail Better Information Services) initiative, is conducting a review and implementation plan to improve, enhance and transform data collection services.
“We are considering how we could put some of our measurement technology on other measuring platforms,” said Goddard. “Also, as the railway becomes more congested with more trains on it, the requirement of data will increase so we will have to have relationships with our suppliers to allow us to flex that demand and not have the static arrangement we have at the moment.
“The service will become unsustainable in the future unless we do something about it. The enhancement programme has set out to look at this.”
Last year, Network Rail launched a strategy for data collection services for the next 10 years, in three parts: Structured Continuous Improvements; Enhance; and Transform.
The first period of structured improvements, which includes short-term innovation cycles to improve the way Network Rail introduces technology to the business; integrated planning; and reducing operating costs; are due to be delivered by the middle of 2016.
During the rest of CP5, enhancements to the data collection services operation will include the automation of train planning and data processing and validation, with remote monitoring of train systems.
“We are also looking to update some of the Network Rail standards, and in the strategy the ultimate goal is to transform the existing monitoring service into a new operation where we’ve got new data quality,” said Goddard.
By the end of CP6, Network Rail wants improved data quality through the use of a consolidated, universal set of interoperable IM trains, which can go anywhere on the network with high-speed measurement migrated to service trains.
“We will still need to have a measurement fleet in the future, but it will be smaller and cleaner, which can traverse the entire network,” said Goddard. “Whatever changes we make to the data collection service we need to make sure we carry on providing accurate data to the business and don’t compromise the safety of the railway.”
Earlier this year, Network Rail requested information on innovative data collection technologies and capabilities that could be introduced into a railway environment to meet its future data collection requirements.
“That’s what the programme is doing,” said Goddard. “We will be following up on our request for information shortly and going to market to tender for a framework where the service is integrated to transform the service in the future.”
There are two immediate problems with the idea of transferring the technology onto passenger trains.
“First, they don’t necessarily have the room within the locos or elsewhere to accommodate the size of equipment we are on about,” he said. “The bigger problem though, of course, is that with the IM fleet we cover the whole of the network and if we transition to passenger trains, they will give us some of the data but there would still be large parts of the network we’d have to measure. On the upside, we’d get more data frequently, but on the downside we wouldn’t get the whole network coverage.”
He added that Network Rail is considering the feasibility of moving equipment from the IM fleet over to other forms of collection platforms, and will look at whether there is a business case for it. “Technically we know it can be done, as we’ve run a pilot on some Virgin trains that have collected track geometry data,” said Goddard. “But what no one has really addressed in sufficient detail is the end-to-end value chain of collecting that data, getting it off the train, processing it and giving it to the customer.”