It seems every industry is awash with all things digital. Rail is no exception. But within this heavily-hyped term there is real tangible benefit, we’ve just got to be prepared to think differently – and in some cases wait a while. In the case of private sector operator and maintainers, lengthy franchise periods and concessions compound the ability to change. So how do we plan our future, instead of waiting for it to pass us by? In this article I will look at technology for depots and maintenance facilities and how it can support future operational models.
Some things don’t change. We still need to be safe at work. We still need to move with the times and our passenger’s needs. And we still need to find better ways of doing things. Over these pages I have provided a view of these three things so we can understand how technology will interact with them.
Depots can be fraught with hazards. Moving vehicles, maintenance pits and gantries, large machinery, hazardous substances, electricity to name but a few… The latest industry figures state that last year yards, depots and sidings alone accounted for 21% of the total harm to the UK rail workforce.
Fleets continue to be fitted with new devices or onboard system upgrades to aid their operation, giving maintenance teams a constant battle to keep up.
Many depots still heavily rely on paper. Systems need upgrading or replacing. Staff must get used to new working practices and, most importantly, these have to be better and more cost effective than before. On average at least 40% of any asset owner or operator’s cost is maintenance.
Then there is that broad term, ‘technology’. We have computational technology like your home computer (more likely to be an iPad these days), and operational technology such as onboard systems, traffic management or devices specific to an asset or system. There are also building-block technologies between, such as the Cloud, security, applications and databases. The list is long and we shall be touching on these as we delve deeper.
Where do we start?
To begin we should wander around the minds of some organisations and people who are pushing the boundaries. To help filter, simplify and put this into context, let’s start with why depots exist: ‘To get vehicles in and out as quickly as possible.’
Simple, right? Wrong. Getting a vehicle into the depot safely and onto the right road is a skill. Today, it requires a lot of human to- human communication, experience and complex planning on whiteboards. Even mature technologies like GPS, despite being prevalent in most moving assets today, are only really accurate enough to say when the vehicle is due to arrive. It’s not so great at saying precisely where they are in the depot at any point in time. Then there is the need to gather parts, tools and technicians. To drive efficiencies while staying safe, you need all these things to arrive synchronously.
The consumer world has seen an explosion of apps and devices to capture, monitor or control mechanical or software systems – collectively known as the Internet of Things (IoT). These ‘Things’ have huge potential in addressing issues with ageing fleets and in-service performance, but before we start bolting things on there are common themes to consider.
Most enabling IoT technologies (building block stuff) are complex, costly and require robust planning to cope with the waves of data they capture over the years they operate. But systems are only one side of the coin. To realise tangible savings you must think, act and operate differently. It is estimated that IoT is 5-10 years away from implementation in most businesses, beyond pilot schemes, because of the impact to operating models.
Communication between systems can be proprietary and open up new rules around data management and stewardship. It is these themes and the requirement to justify a return on investment that holds us back.
Products are already on the market that provide forward notification and communication between traffic management and the depot. Mainly badged as ‘protection systems’ and focused around safety, these are just one element of the overall solution. Typically they are big investments and the depot is built around them in order for the technology, signalling and automated operation to join up.
So whether you are starting from scratch, radically changing your operating model, or making minor improvements using IoT, we should consider two initial Cs: Capacity and Connection.
Capacity is both the amount of data that can be processed at any one time and the bandwidth needed to transmit it securely. Connection is about speed and coverage.
The demand for capacity in the consumer world and a need to always be ‘connected’ continues to drive device demand. In business, this expectation pushes people’s thinking. Phrases such as “How can I watch a whole series of Game of Thrones on my iPad sat in a coffee shop, but can’t locate my vehicles on the depot?” are common refrains.
Lots of things are at play here but they ultimately come down to the two Cs. TV streaming services have phenomenal systems and processing power at their command, while compression technology to squeeze the most out of bandwidth has advanced significantly. Mobile networks, in particular, constantly evolve to keep up with demand. Sadly, some procurement contracts and suppliers don’t run at the same pace.
The next big advance will be 5G. Throughputs of 20gbps downstream and 1gbps upstream, all at low latency (5ms), have the scale to cope with huge amounts of data – perfect for IoT solutions and high-definition video. The latter will enable vehicles to act as their own intelligent ‘third eye’, transmitting video in milliseconds and making decisions about obstructions. This is important in depot environments where access doors or staff can be in the way of vehicle movements. Drones with infrared cameras and HD video can also play a part in the safety and security of the depot using these networks.
Research suggests that ultra-reliable low latency networks will be in place by 2025, so any major procurement and investment made now should consider how easy switching over and taking advantage of these types of networks will be. Remember, these solutions need to be as close to real-time as possible and accurate to within millimetres. This means lots of data being streamed and transmitted securely in milliseconds. Seven years is not far away… think how quickly your mobile phone became smart.
When physical and digital collide
Now our vehicle is on the right road and we are confident we can locate it in real-time through our reliable mobile network, what attention does it need?
Condition Monitoring has been around in the aviation and automotive industries for years. Direct connections to onboard systems extract data that is converted into service intervals based on how the vehicle performs and how it is operated.
Within rail this concept has been around for a while, but has never really been adopted at scale outside large rolling stock manufacturers. IoT-based pilots are therefore the logical step to gauge how this data will positively inform operations. But for it to be a true game-changer, you will need an environment called a Digital Twin – a virtual, digital counterpart of a real, physical object.
The Digital Twin enables software and systems to interact with it rather than the real object, improving maintenance, upgrades, repairs and operation of its physical equivalent.
In order to be truly useful, the twin needs:
Once again, we see how important it is to understand capacity and connection. But here a third C comes into play: Compliance.
Digital twins generate and feed off colossal amounts of data. Understanding the flow of this data, who owns it, how trustworthy it is and how much it will cost to obtain goes some way to explaining why only 5% of such objects are virtually modelled today. This statistic covers all major engineering-based industries, so there is a long way to go. As time goes by, the data platforms, computing power and bandwidth will get bigger, faster and cheaper, meaning this will be a key contributor to how physical environments are maintained.
Culturally – arguably the most important C – this is a seismic shift, not only in how we do things, but also who does them. The skills and Capability (a fifth C) found in maintenance teams will need to move closer to both digital experts and engineers, coming together to drive predictions.
This change will open the door further to technologies such as Augmented Reality (AR). AR is the enhancement of images of real-life objects by analysing what the camera sees and feeding state and other contextual information to the viewer.
Technology giants are already launching AR platforms; Google Lens (a suite of Artificial Intelligence visual search technologies), for example, can overlay interactive information on what a device’s camera is viewing. In engineering terms, this could be what the part is, its model number, engineering change data and stock-holding location, as well as contextual detail about the vehicle’s recent journeys. To get this today you need to interrogate multiple systems and people and may still not get the answer you are looking for.
Digital printing is the next obvious step. Creating replacement parts needs detailed information that the Twin would already hold; spares would be fabricated on-demand based upon predictions and when the vehicle is due to arrive. Again, large maintenance providers and rolling stock manufacturers are investing significantly in facilities dedicated to 3D printing (also known as Additive Layer Manufacturing). Safety and compliance is the main restraint before we start fitting digitallyprinted parts to our vehicles, but in under five years this will start to become reality.
There is even embryonic research into 4D printing (4DP), allowing the printed object to change shape. Smart materials will start to solve design and engineering problems that arise from the limitations of current materials, although precision engineering applications of 4DP mean we will not see any major adoption for at least 15 years.
Are our days numbered?
In order to answer this we need to consider what works for our business, people and assets/vehicles. What is true is that in the future we cannot solely depend on people. We must trust our data more, and in particular that related to what our vehicles are really doing when in service.
Data is critical. We cannot escape it so we must learn to manage and use it better.
One thing is absolute in all the research – until we develop an interface into the human brain we will always need experts to analyse and reason with it. The combination of the two and the execution of a new operating model is where organisations will succeed.
But the human element can’t rest on its laurels. DARPA (the US Government’s Defence Advanced Research Projects Agency) has significant funding invested in developing the Brain Computer Interface. This is a 1cm3 chip implanted into the brain that will send up to a million channels worth of neurological brain activity to electronic devices, such as prosthetic limbs. It is expected to extend to hand gestures that will both communicate and manipulate physical and augmented interfaces at the same time.
But safety first… I predict you can safely wave goodbye to your colleagues for at least another 30 years without causing an accident!
Adam Collins has worked for and with some of the world’s largest technology companies, as well as running his own business. Over 23 years he has consulted, led global change programmes and managed large technology portfolios in Government, utilities, retail and now rail.
In 2015 Adam Joined SNC-Lavalin to lead its Digital Services team which consults, develops and supports clients in their adoption of technology. Hear more from Adam at this year’s UK Light Rail Conference on 17-18 July www.mainspring.co.uk/uk-light-rail-conference
Share this news:
Leave a message: