By Chris Warrington and Matt Keepman of Tracsis
Our industry is not as unique as we think it is. Sure, there are unique aspects to the operational models but the transformational challenges we face as we adapt to new technologies are the same ones faced by a range of sectors. Maybe there’s something we can learn from them.
Railroaders don’t have all the answers and the models that are currently being perpetuated are not going to solve the problems. At best, we’re being incremental. Other industries which are more mature in terms of technological evolution are taking a wider, holistic view of data value and management: they are joining up systems to enable the delivery of more useful data.
Data needs to be considered in the context of the broader network because network efficiency is only as effective as its weakest link. For example, in the utilities space, electrical infrastructure companies are embedding sensors on transformers across their networks to measure variables such as humidity and temperature so they can better predict when transformers will fail. The data may predict a transformer will fail in three to six months – but if it’s on a critical part of the network and a truck is in that area, why not fix it sooner?
Another example can be found in the airport industry, which is ahead of the curve in adopting new processes and ways of managing data infrastructure to improve efficiency, particularly in the UK and Europe.
Airports have significantly altered their business models within the past decade as privatization and increased private equity investment has driven a push towards higher efficiency, more capacity and improved customer experience.
Old processes were no longer going to cut it. A different mindset was needed, so airports began recruiting CEOs and operational executives from non-aviation backgrounds to provide fresh perspectives and ideas (Hello CSX and Joe Hinrichs), augmenting the knowledge and expertise of decision-makers with decades of airport experience.
Airport procurement has also been evolving to be more open and collaborative, especially in the early stages of the procurement process. For example, airport management may get the ball rolling by sharing their top strategic priorities with suppliers prior to requesting ideas and strategies for achieving those outcomes. Instead of launching a prescriptive tender process based around a list of requirements, the scope is focused on a list of outcomes. Working collaboratively to solve problems has been a game-changer for both airport management and suppliers, resulting in better integration and efficiency in the supply chain.
An openness to different ways of thinking has led to more effective methods of leveraging airport data pools. Rather than relying on fragmented, siloed systems – one set of data for on-time arrivals and departures, another for customer throughput, another for luggage information, and so on – enterprise data is increasingly being treated as a singular asset, with somebody accountable for design and management.
Various operational systems are integrated into a unified data platform – the Airport Operations Database (AODB) – to provide a ‘single pane of glass’; an interface that shows at a glance the overall health of an airport. The underlying thinking is that it’s rare that an isolated decision in one part of the operation impacts just one asset; ripple effects are the rule rather than the exception.
There has also been a drive towards “management by exception”; rather than drilling down into every routine activity, operational systems focus on significant deviations from the norm in processes or data. This means that operators can focus on what is important and/or urgent.
How can rail networks optimize data infrastructure and improve processes?
The rail supply chain is populated with highly specialized businesses, so it’s unsurprising that both vendor partnerships and data pools are fragmented. Deeply ingrained risk aversion and an underlying belief that only rail people can fix rail problems creates a situation where rail data is not being harnessed to anywhere near its full potential.
Historically, cost has been the biggest barrier to data infrastructure investment, but that is decreasing as AI and IoT technologies evolve and advance. Yet, railroads remain slow to get on board. Less than 5% of rail cars in the United States are fitted with telematics, despite the efficiency and performance benefits being well known for years.
A first step for rail companies may be to take a page out of the airport playbook by adopting a collaborative, partnership procurement model, prioritizing outcomes over process. Currently, organizations are getting what they ask for but it’s not necessarily what they need: to elevate beyond a risk-averse, prescriptive approach that restricts innovation to an outcomes-based partnership that evolves the capabilities of the technology. Railroads should better define their strategic priorities and collaborate with vendors to achieve these goals with an openness to new ideas and solutions. While partnership means different things to different people, fundamentally it should be about railroads and vendors working together towards the same vision.
Secondly, what railroads need right now is a single system of record, a single data platform, that consolidates all the data pools, cleans up the data, and synchronizes it. Integrating data from various sources will provide a holistic view and ensure that all parts within a rail network are operating efficiently and effectively.
There has to be a better understanding that static data without context has limited value. For example, data collected from a set of points on a commuter track between two cities may exhibit the same performance criteria and outputs as data from a set of points on a freight mainline linking two small rural communities – but failure on the city track is inevitably more consequential.
Moreover, a significant amount of rail data collection around asset condition is based on a single point in time – there’s no predictive element so the data has limited value. We need to evolve that proposition by taking the data and synchronizing it, enhancing it with data based on variables such as temperature, precipitation and other key factors to expand the data set and deliver a more accurate root cause analysis of the problem.
Taking a holistic approach
In short, the North American rail industry needs to adopt a holistic approach to data infrastructure while fostering more collaborative vendor partnerships to get the best return on technology investment. This will require more than a shift of mindset and a leap of faith; it will require a mandate from the C-Suites of our railroads and buy-in from key decision-makers. It may also require executive recruitment beyond the usual hunting grounds.
Chris Warrington is the Managing Director, M&A and Innovation, at Tracsis, a leading international provider of software-based rail solutions for more than 20 years. Based in the UK, Chris has previously worked in the airport and utilities sectors.
Matt Keepman is Vice President of Sales – Freight, for Tracsis US. He brings over two decades of experience in managing Class I and other strategic accounts within North America’s rail industry.