25 July 2017
In June, I had the privilege of visiting Australia for the first time. A wonderful country that, in addition to having some of the friendliest people in the world, is home to progressive cities such as Melbourne and Sydney, which understand that transportation is at the heart of our societies.
For cities, infrastructure providers and governments, the critical question that drives most decisions is, “How do you ensure that passengers waiting for a train are getting a train on time?” Beyond the rolling stock and infrastructure, the answer lies in data specialists who monitor and analyse the billions of data points that a train sends out every year.
But data itself is worthless unless we have the knowledge and capacity to turn it into information. For the rail industry, that means making decisions about whether to take a train out of service to be repaired or keep it running for a few more days. Decisions such as analysing data and information on what parts of the train have broken, what spare parts have been used and what is still available, what geographic regions it has travelled through (is there a hill that is notorious for causing problems?), and whether it’s near a service depot that has capacity to provide maintenance. This level of analysis is required to turn raw data into valuable information that impacts operational capabilities.
So what are some of the key pillars of data and digitalization in the rail industry, and more importantly, what is the potential?
The first step to consider is the availability of the vehicles. If you do not know the condition your vehicles are in, you do not know when one of them will fail. This knowledge is a basic prerequisite for the optimization of operation control - the actual production process in rail transport.
Risk detection and maintenance has since gained significant ground, at Siemens and across the industry. Predictive maintenance is essential in improving availability of rail assets, as it allows you to understand a problem before it arises, replacing the component at a time when it does not disturb operations, reducing the risk of unforeseen events during a journey.
New data and new possibilities
The digitization of the operation control is a challenge, particularly in large fleets and long distances, because the dependencies are significantly higher. The more projects we implement, the more data we can generate and analyse - especially as we work on projects worldwide where the length of the routes and trains vary as well as the climatic conditions or the capacity utilization. Therefore, with each new challenge, we as an industry have the opportunity to gain new data and insights that we can use for other projects as well as predictive maintenance.
A great example is how a small team of Australian-based Siemens rail engineers designed and developed a Remote Diagnostic and Advisory System (RDAS) – which is now part of the global Siemens Railigient platform – that for the first time, brought together disparate systems on the network into an integrated platform, giving operators the ability to view all their assets in real-time through one consolidated application.
Cooperation that drives continuous development
For the rail industry to flourish in the digital world, it is important that we not act in silos but work together. An interesting example is in Russia, where we operate not only a center for data analysis, but also in a joint venture, manufacture heavy locomotives. It has opened up a whole new data source for us, from which we can also learn for other global projects.
Around Moscow, there is a dense network of regional trains and suburban trains, where currently about 40 trains run in on a ring route around the centre.. From 2018, the trains should drive autonomously. This is, of course, a major challenge in that there should be no surprises during operations. We need to understand the interactions even better, and we are currently working with our joint venture in our Russian data analysis center.
Unleashing the full extent of data
Today we are relatively at the beginning of a gigantic learning process, which will open up many possibilities. However, the speed of this process is also very impressive, so we can probably expect very fast development and networking.
In Spain, we worked with Renfe, the national railway, to boost on-time operations on the high-speed rail line between Madrid and Barcelona. The rail trip takes two and a half hours, and it competes with flights of an hour and twenty minutes. When the trains started, 80% of travellers took the plane, and 20% took the train. As of early 2016, those numbers have reversed: 80% are taking the train, and just 20% are flying. That is because those trains are punctual.
With each new project, we get new data from which we learn, and in turn, we get benefits for existing projects. It is also interesting in this process that thanks to the merger of vehicle and maintenance data, we can now make even better predictions. And this is still an important basis for our work.
All of this is important, because with the insights we can create from data we can help our customers gain more value from their assets, ensure better operations and ultimately help them be more successful in their business.
Gerhard Kress is Global Head of Data Services at Siemens Mobility. An extended version of this article was published in Railway Digest – July 2017 edition.