HEAVY-haul railways are regarded as one of the bastions of railway technology. The desire to carry higher quantities of commodities from mines to ports and industrial sites has compelled railway engineers to come up with innovative solutions that have continually pushed the limits of what is possible.

Throughout its 60 years of publication, IRJ has charted the key developments, many of which have filtered into conventional railways. In track infrastructure, the growing use of continuous welded rail, direct fixation fastenings and high-grade steels for rails and wheels are among the key developments. We have also featured innovations in rail monitoring using track geometry cars, enhanced rail flaw detection and rail profile grinding, which have helped to improve track performance and reliability. Concrete sleepers have also played a crucial role in providing additional stability as trains have become heavier.

Heavier trains demand effective traction, and the development of ac locomotives in the 1980s is considered a key step in this process, enabling operators to cut the number of units required, in some cases by half. Likewise, distributed power has facilitated longer and heavier trains, along with the use of ECP braking, and permanently coupled wagon groups.

The development of longitudinal drive simulation for driver training is also considered critical to enable drivers to handle train lengths of 2.4km and above. Evolutions in Centralised Traffic Control (CTC) have similarly helped to boost network capacity and safety. Finally, automated operation beginning with single driver and now driverless trains pioneered by Rio Tinto in Australia with Autohaul, along with automatic track inspection systems and wagon unloading, hints at the potential of further automation that will help to transport even larger volumes.

For Ms Lisa Stabler, president of Transportation Technology Center Inc (TTCI), the Association of American Railroads’ (AAR) test facility located in Pueblo, Colorado, the improvement in safety on the North American network since the turn of the century is the major milestone which demonstrates the sector’s progress.

“If you look at the statistics provided by AAR, employee injuries are down by 46% since 2000 while train accidents are down by over 30%,” Stabler says. “That really comes down to the focus the industry has that safety is the number one priority.”

The other major milestone for Stabler is the increase from 120 metric tonnes to 130 tonnes per wagon for unit trains and the establishment of a coast-to-coast, cross-border interoperable heavy haul network in North America, which now accounts for 2.7 trillion tonne-km every year. “We are allowed to interchange 130 tonne wagons and we do so freely every day,” Stabler says. “It is now done so freely, that the only thing stopping traffic is the necessity for border exchange.”

“Employee injuries are down by 46% since 2000 while train accidents are down by over 30%. That really comes down to the focus the industry has that safety is the number one priority.”

Lisa Stabler, president of Transportation Technology Center Inc (TTCI), the Association of American Railroads’ (AAR) test facility

In Australia, and particularly in the Pilbara region, rather than an interoperable network of multiple railways, heavy-haul operation is largely on point-to-point lines. Mining companies such as Rio Tinto, BHP and Fortescue have pushed for optimisation in signalling, wheel profiles, and running methods in order to steadily increase axleloads to beyond 40 tonnes, maximise throughput, and carry more and more natural resources extracted from the Australian earth.

The ability to push the boundaries reflects the railways’ growing confidence and understanding of their infrastructure and assets. Enhanced monitoring capabilities makes this possible. And according to Mr Michael Roney, who after a 32-year career with Canadian Pacific (CP) is now a consultant working with heavy-haul clients around the world, is helping to fulfil the mantra of the International Heavy Haul Association: longer, heavier, faster.

Rio Tinto inaugurated the world’s first driverless heavy-haul freight railway in 2019 and other railways now hope to follow its lead. Photo: Rio Tinto.

The willingness to identify, test and implement new technologies is central to the sector’s evolution. TTCI plays a critical role in North America by offering the space for trials to support the Class 1s and other railways. In particular, the 4.3km Facility for Accelerated Service Testing (Fast) test track at TTCI’s headquarters in Pueblo, Colorado, which supports 143 metric tonne wagons and generates about 25 million gross tonnes per month, is a testbed for a range of innovations to see if they are robust enough to enter 130 tonne traffic. TTCI also has agreements with four Class 1s to test in revenue service.

“Fast has enabled us to come up with innovations in steel for wheels and rails and helped us to validate new designs for tracks, bridges and rolling stock, and all of the associated components,” Stabler says. “We also have computational modelling. The suite of products we have for heavy haul - and it also works for passenger rail - allows us first to model the impact of changes, and then test them at Fast to ensure that what we have modelled is correct. We can then take those products and services into revenue service. Computational modelling enables us to run through the different design iterations very quickly.”


In Australia the railways themselves are the test bed. Professor Colin Cole, director of the Centre of Railway Engineering at Central Queensland University, who has 37 years’ experience of working with the Australian rail industry, says heavy-haul operators have embraced experimentation with latest technologies, many of which have gone on to enter operation. He says the high production rates of these railways, along with an emphasis on condition monitoring, facilitates a relatively fast process of evaluation of new technologies and practices.

While progress is commendable, there have been some big ideas over the years that didn’t quite materialise. Roney points to the Advanced Railroad Electronic System (Ares), a forward-looking satellite-based train control system pioneered by Burlington Northern in the 1980s, as a promising solution that didn’t take off due to business reasons, much to the disappointment of many at the time. Another is the failure to develop continuous monitoring of rail stresses. Roney says there have been multiple false starts of technologies that look like they could do it, but ultimately failed.

In a similar vein, there have been numerous attempts to replace ultrasound as a method of inspecting rails, including pulse laser emission of ultrasound and electro-magnetic acoustic transducers.

“I did a survey on this about a year ago and found a number of things that have been tried but haven’t been able to replace ultrasonic rail flaw detection,” Roney says. “There are still parts of the rail which are not seen ultrasonically, such as the rail foot and the bottom area of the railhead. The good news is that it has improved. Artificial intelligence is now helping to interpret those signals, but there is still a need for a new technology that floods the entire rail.”

More controversially, Roney says steerable bogies are one concept that have failed to live up to expectations. While they have gained traction in South Africa, where they were pioneered, the added maintenance costs and tight wheel profiling measurements have put others off. “They should be a godsend for any railway working with tight curvature conditions, but somehow we haven’t got there,” Roney says.

There are also instances where accepted hypotheses were proven to be wide of the mark. Stabler says 20-30 years ago the sector was convinced that just making steel cleaner would get rid of inclusions, porosity and solve the problem of broken wheels. While the steel industry has become a lot cleaner since then, she says that work has since shown that as long as cleanliness characteristics are met, further improvements will not result in additional upturns in performance. “What we need to look at now is alloying,” she says.

Mr Gary Fry, senior assistant vice-president, research and development, at TTCI, says a lot of the solutions are not intuitive, and it is just as important to pay attention to unexpected results as those which researchers expect. “We learn the most when unexpected things happen,” Fry says. “Whether it’s unexpected and ‘wow, that was a lot better than I thought,’ or unexpected in that we thought it was going to be better.

“We don’t just study the components that have a lower performance level - if we have welds in track at Fast that don’t last that long, we study those, but we also study those that last a long time, those at the other tail of the distribution, so that we can compare the features and characteristics to see what made the difference.”

While heavy haul has made significant progress in the last six decades, a number of challenges and opportunities remain. Roney identified what he considers the five major challenges facing global heavy-haul railways in a recent presentation to the US Transportation Research Board:

  • deriving more value from already deployed assets and technology
  • improving operation efficiency and customer experience
  • navigating the train in the fourth industrial revolution
  • providing new rail capacity in a cost-effective manner, and
  • continuing to strive for zero derailments and zero injuries.

Like it has throughout the last 60 years and even before that, the heavy-haul sector will continue to rely on advances in technology to overcome these challenges. Data is the oil of the fourth industrial revolution, and railways are increasingly deriving data from enhanced monitoring of more and more assets in order to better understand and improve performance.

This is offering accurate projections of asset needs ranging from wheel demands to rail renewal cycles and is translating into targeted maintenance spending. It is also, as Roney puts it, “turning finders into fixers,” by sending track or vehicle maintainers to specific locations. This helps to reduce manual inspections of track and vehicles, helping to improve worker safety and limit dwell times, a key performance indicator for all freight railways.

Stabler says intensive efforts to harness asset performance information and operational data began in North America in the early 2000s. However, the information available to a single Class 1 at that time presented an incomplete picture.

Marrying information from multiple railways dramatically enhanced this process. AAR’s Railinc subsidiary is now playing a critical role by offering a centralised database of information retrieved from multiple railways, including real-time wagon movements and Positive Train Control (PTC) system event log files.

The North American Class 1 railways have created a cross-border interoperable heavy-haul network that accounts for 2.7 trillion tonne-km every year. Photo: David Gubler

A variety of products are available and the wagon data in particular are helping financially by offering a clear picture of where a specific wagon is and identifying who pays for a particular asset as it goes from journey to journey. Railinc is also increasingly helping the railways operationally by offering insight on wagon measurements and performance metrics as well as forecasts of traffic conditions to prevent or reduce congestion.

Track monitoring

Fry says in general the North American industry is now very data driven with railways actively employing information technology teams which include people with degrees in data science. And while wagon tracking capability is now strong, he says further improvements are expected in infrastructure monitoring.

TTCI is supporting this process by developing decision support tools for the railways to facilitate predictive track and infrastructure maintenance by identifying imperfections before they become defects.

This process is based on developing predictive models that assess data streams retrieved from asset sensors and monitors which identify a variety of event signatures using standard predictive analytics techniques. Among the monitoring technologies available is ultrasonic inspection, which measures the running surface condition of rails and wheels. Other techniques are also in use to facilitate subsurface inspection of rails, sleepers, fasteners and the ballast layer. And crucially, rather than working in isolation, networks of sensors are providing information on trends, offering railways a more complete picture of overall network performance.

“The better data they have, the safer the railroads become,” Fry says. “Instead of saying I am going to spend a million dollars in this zone of track, period, they can now weigh up where this money is best spent. Being data driven means these dollars go to where they actually need to go.”

TTCI’s work on enhanced detection techniques comes hand-in-hand with material development projects for wheels, rails and welds, new post weld treatment to improve fatigue performance as well as developments in bridge and other civil engineering design. “We have 35 different initiatives, but they’re all focused on safety and enhanced performance,” Fry says.

Future improvements are expected through research to further enhance the understanding of the wheel-rail interface, specifically how to control and understand what happens in the contact patch location. There is also work underway on vehicle dynamics to offer an improved appreciation of how a vehicle will respond to different track and track structure conditions. Finally, Stabler expects advances in automation to facilitate in-motion inspection of both vehicles and infrastructure.

“If you can thoroughly understand what happens to a wagon while it is going 100km/h, that is much less disruptive to the industry than having to take those wagons out of service to do static measurements,” Stabler says. “In some cases, the static measurements are not necessarily predictive of what happens when you have a dynamic measurement. The concept of having dynamic measurements for everything associated with rolling stock is the next goal.”


Traction, specifically the fuel source, is another major challenge for the future. While heavy-haul networks in South Africa, Sweden, Norway, China and many lines in Russia are electrified, in North America, Brazil and Australia, diesel remains king.

Diesel trains are cleaner comparatively with road transport. However, with larger companies recognising that they must move with the times and reduce their carbon footprint, the pressure is set to increase to find alternatives and reduce emissions.

This is no easy task, particularly in remote areas such as the Pilbara. “You would have to build not just an alternative power system, but an entire supply chain,” Cole says. “We have got plenty of sunshine, there is no doubt that we can get the power, and it is something that will have to be done in the next couple of decades. I think we’ll see heavy-haul railways with their own private power stations.”

In North America, trials are underway with alternatives such as LNG, LPG and bio-fuels, and it is likely that battery and hydrogen fuel cell applications could emerge soon in yard settings. However, the reluctance to electrify persists despite the benefits of reduced headways and improved control as well as reduced emissions - Transport Canada estimated in 2019 that electrification of CP and CN’s mainlines between Montreal and Vancouver would eliminate 43% of Canada’s rail freight emissions while producing $C 520m in annual savings.

“The railroads are not going to buy a system just because they say they’re quantum systems.” They’re going to buy whatever works the best.”

Gary Fry, senior assistant vice-president, research and development, at TTCI,

“We looked at electrification at CP around 30 to 40 years ago, and actually built some sample catenaries to see how they stood up in the mountains,” Roney says. “We did an economic study, but at the end of the day the big worry was that ultimately shareholders would not be happy that their dividends would be reduced for a number of years… it is much easier for state-owned enterprises to do electrification.”

As road explores possible electrification, the pressure is likely to continue to build. It might be that one railway eventually takes the plunge and others follow suit, although this is by no means certain.

Automated trucking is another area of concern for railways. The Class 1s are exploring how they might introduce automated processes to improve throughput and the efficiency of operation by harnessing PTC infrastructure now at their disposal.

Rio Tinto has of course pioneered driverless operation and other Pilbara railways are exploring the technology. Brazilian railways are also understood to be very interested in its capabilities. “I would be very surprised if more Pilbara railways do not move to adopt ATO within the next 10 years,” Cole says. “In any case, they will be early adopters of new train control technologies.”

Railways have traditionally been perceived as relatively slow adopters of new technology. Roney says it is always a worry of his that things happen in the road and rubber tyre business faster than they do in the rail sector. For instance, he says the fact that railways buy locomotives and wagons with the expectation that they will last for 30 years is a problem. He says road also traditionally spends more on R&D than rail.

Yet Roney is encouraged by the heavy haul sector’s efforts to digitalise and embrace a new set of opportunities. “I think that things are going to pick up in pace,” he says. “And as we continue to mine the digital world, things will happen faster.”

Stabler says from her experience of 20 years in the automotive sector before shifting to rail, the perception of the rail industry as “stodgy” and “resistant to change” is wide of the mark. “It embraces those technologies that are needed to help improve safety, reliability and efficiency,” she says. Fry agrees. He says the sector will continue to look at and adopt techniques and processes that offer an improvement in performance, safety and reliability. This includes Artificial Intelligence, which is already finding its way into rail, and quantum computing.

“The railroads are not going to buy a system just because they say they’re quantum systems,” Fry says. “They’re going to buy whatever works the best. It’s an issue of pragmatism, competitiveness, and economics.”

The prizes for getting this right - higher productivity and improved performance - are self-explanatory. And as freight railways continue to seek increases in throughput, improved efficiency and enhanced safety, the desire to explore and use new technologies will remain. The heavy haul sector has witnessed significant progress in the past 60 years, and this looks set to continue for the foreseeable future.