Global demand for transportation has exceeded the available capacity. In today’s context of climate and pollution concerns, it is no longer possible to meet demands by increasing the physical capacity of transportation networks. Instead, we need more innovative solutions – and this can be enabled by technology.
The railway industry has embraced technology at a slower pace than other modes of transportation. Across most of the Americas, Europe, and Asia, planning and operating trains is still a human-intensive process. So the question is: can advanced control and automation deliver substantial benefits to railway operators? Historically, many transportation systems have been human intensive, but are benefitting from an amazing level of automation today
Control and automation in transportation systems
About 100 years ago, cars were steered using a tiller, and speed controlled by a hand-held throttle. The starter, carburetor, choke valve, were all manually operated. Today, most of these functions have been automated. With autonomous cars, almost incredibly, driving itself has been automated. As these cars become mainstream, they will free up the chauffeur space (that is an increase in throughput!) and also open our minds to the amazing possibilities of automation.
As we understand the cause-effect relationship between the parts of a complex system, we can automate their operation. Railways are ready for such an evolution.
Getting railways on track
Increasing railway infrastructure capacity is an expensive proposition, and doesn’t guarantee proportional increases in efficiency. Despite a 75 billion dollar investment in railways in the US between 2009 and 2013, the American Institute of Civil Engineers assessed railway infrastructure and graded it C+. A much cheaper alternative is to supplement existing infrastructure with advanced sense-analyze-control algorithms. These methods process vast amounts of data and deliver superior performance to that of humans, who can only process a limited amount of information at one time. Precise tracking and prediction of train movement allows them to run faster, with fewer stops and more frequent trains, thus increasing the effective capacity of existing infrastructure.
In India, we know that the current dispatching methods draw upon over 100 years of organizational experience, and are as efficient as humanly possible – but the key word here is ‘humanly’. The beauty of automation lies in always finding that last drop of performance that may be missed by an inexperienced, distracted, or tired human dispatcher. Our calculations show that extracting just 5-7% additional throughput from Indian Railways may be worth as much as 10,000 crore rupees in additional revenue.
How, exactly, does sense-analyze-control work?
Let us consider an example. The Kharagpur division of Indian Railways consists of approximately 2,000 kilometers of track with 150 stations and handles close to 300 trains a day. The division is split into five parts, each controlled by a human dispatcher who allots tracks to each train. Decisions to reorder trains because of emergent delays are driven by several conflicting factors, including train priorities, downstream conflicts with other trains, freight loads, and passenger connections. It is difficult for a human to process all of these objectives simultaneously, and to come up with a solution that is optimal for the network as a whole.
A ‘sense-analyze-respond’ system works well in such scenarios. The ‘sense’ aspect fuses a vast amount of data about the status of trains in the network, the ‘analyze’ aspect considers the implications of each available choice, and the ‘respond’ aspect allots various track resources to each train, taking into account physical capabilities and safety rules. The advantages of such an approach are:
- A single system can reduce the workload of several dispatchers
- It improves the quality of the solution (for example, the system can find out what is happening in the entire Kharagpur division, as opposed to one small area
- It is faster and more accurate than manual decision-making, which leads to safer, on-time transport
The use of Artificial General Intelligence in Railways
The development of Artificial General Intelligence allows automated systems to learn from historical operations, making them more efficient as time progresses. This means the train dispatching system can observe past scheduling decisions and outcomes to make better decisions: for example, sometimes allowing Train A to go before Train B, and at other times allowing Train B to go before Train A. Such new techniques allow us to detect the long-term repercussions of short-term decisions, enabling more efficient and yet more robust operations.