SIEMENS

Air traffic is booming all over the world, and passenger miles are growing by five to six percent every year. In New Delhi, the third terminal of Indira Gandhi Airport was opened in 2010. In Frankfurt, Germany’s biggest transportation hub, a fourth runway opened in 2011. And in the U.S., investment in “NextGen” advanced airport infrastructures and runways is continuing at a steady clip. To take just one example, the international airport in Los Angeles, which is the sixth-largest in the world and serves 60 million passengers annually, is being significantly expanded. The mobility of people and goods is also increasing on roads, rails, and ocean routes. The International Transport Forum estimates that global passenger traffic in all categories of transport will triple or even quadruple in size between 2000 and 2050 and that the transport of goods will grow by a factor of 2.5 to 3.5 during the same period.

This tremendous increase in traffic can be most clearly felt in large cities, which are the recipients as well as the victims of this development. Because they are responsible for about 80 percent of the world’s economic output, cities are the hubs of the global mobility network for people and goods. Just as with the exchange of data on the Internet, hubs or nodes are crucial to maintaining the stability of the network as a whole. Cities have realized this fact. In a survey commissioned by Siemens several years ago, more than 500 mayors and urban experts all over the world defined traffic infrastructures as being by far the most important factor if their respective cities were to remain attractive as business locations.

Ideas for Future Hubs. Does this mean that road and rail networks need to be continually expanded? The traffic experts who participated in the survey have other priorities. Above all, they want to make better use of the existing infrastructure — a step that is less expensive and more environmentally friendly. This approach is also the focus of future scenarios conceived by Siemens’ Mobility and Logistics Division and the “Future of Hubs” idea competition conducted by the division. “Our employees submitted 140 ideas,” reports project initiator Huschke Diekmann. The new concepts were put on an intranet platform so that colleagues could evaluate and comment on them. The ideas that were judged to be the best all had one thing in common: They agreed that the networking of individual modes of transport has particularly great potential.

The winning project was called “An Intermodal Passenger Information Platform.” It called for the linkage of all the information about all modes of transport within a city in a single software solution. All the schedules of every form of local public transport, as well as the traffic situation on streets, would be made available. This would make it possible to offer an app that could be used not only to plan a route from house to house but also to continually check all the alternative routes in real time and get recommendations about the best one. Siemens presented a prototype of this system in late 2011 at the sixth National IT Summit in Munich. The demonstration presented an “intermodally” traveling businessman who navigates around a traffic jam using a smart phone and the Internet. In the process, he switches from his car to a train and then to an e-scooter and finally reaches his destination on foot.

The contest’s second prize was awarded to an idea for “intermodal goods transport.” On average, it takes up to 12 hours to unload a freight train loaded with containers. The cars must first be decoupled from their electric locomotive and then pushed by a diesel-powered locomotive to a track without an overhead cable. A portal crane then loads the individual containers onto trucks. According to inventors from Siemens, the transshipment process could be performed under the overhead cable using a much simpler and lighter unloading bridge, and the containers could be moved around via a parallel track similar to a conveyor belt. An entire freight train could be unloaded in less than two hours and it could then continue its journey. Using similar technology, containers in seaports could also be unloaded from ships directly onto railroad cars. Smart containers could play a key role in the goods transportation of the future by automatically providing information about their destinations and delivery dates to a logistics IT system.

One Control Center for All. In Germany, a research project known as the Total Airport Management Suite (TAMS), which is funded by the Federal Ministry of Economics and Technology, is designed to exploit the full potential of intermodel systems. Siemens, the project’s leader, has worked closely with the German Aerospace Center, Stuttgart Airport, and other industrial partners on TAMS, which was completed in early 2012.

The basic idea behind TAMS is simple: Link everything. Thus, at an airport, key factors, such as capacity and number of take-offs and landings should be coordinated with flight plans and dozens of related systems, such as the timing of refueling and luggage loading activities, the number of people checking passports, the capacity of gate areas where planes dock, and the destinations of catering trucks. Today, these tasks, as well as others, are generally performed by independent service providers. Each of these providers dispatches its employees according to a coordinated plan, but from its own control point. Like clockwork, each gear wheel connects with another one — until a major disturbance, such as a snow storm, occurs. To date, the IT systems of service providers have been linked at best by a shared database. As a result, alternative plans have to be laboriously coordinated by the heads of operations.

Things are very different in a TAMS automation environment. Here, all service providers are linked in a single control center that coordinates all operations throughout the airport. The IT systems of the individual companies are linked in such a way that their employees are supported by integrated assistance functions when they need to make decisions.

“In airports that are operating close to capacity, TAMS can increase the number of airplane movements per hour by about ten percent,” says Dr. Christoph Meier, who is responsible for airport IT in Siemens’ Mobility and Logistics division. This estimate is based on simulations carried out at the German Aerospace Center (DLR) in Braunschweig. TAMS also has a positive influence on CO₂ emissions. That’s because the integration of air traffic control means that every airplane rolls to its starting point only if it can take off a short time afterward. Lines of airplanes waiting to take off can thus be almost completely eliminated — along with associated fuel use. Because decisions taken by air traffic control, such as changes in the direction of take-off, no longer come as a surprise to other operators at an airport, punctuality simultaneously increases by up to 20 percent. This results in clear economic advantages for airlines. The European air safety authority, Eurocontrol, estimates that the costs that are caused by all flight delays in Europe total as much as €1 billion annually.

The results of the TAMS research project have been so encouraging that Siemens intends to offer TAMS as a product as early as 2012. Airports will then be able to buy the complete software architecture, as well as a control center, from Siemens.

From the passenger’s perspective, it’s crucial not only to land on time but also to reach his / her ultimate destination quickly. However, in many cases, arriving passengers discover that the local infrastructure has failed to keep pace with economic growth. This can become particularly clear if they’re taking a taxi. In many cities, the average speed of a taxi during rush hour is less than 20 kilometers per hour.

Diekmann, the technology strategist, believes there’s no such thing as a sensible standard solution. “Every city is different,” he says. To discover not only the quickest route, but the one with the least environmental impact, Dozens of parameters must be taken into account, including emission figures for different modes of transport and lengths of traffic jams. Solutions depend on the accuracy of data. The data must then be registered and processed by traffic guidance systems such as the ones Siemens has installed in over 1,000 cities.

Siemens commissioned one of the most modern traffic guidance centers in Berlin in 2005. The flow of traffic throughout the city is monitored with the help of video cameras and almost 2,000 sensors, most of which are induction loops embedded in asphalt. More than 1,700 traffic lights and 300 overhead sign gantries can be controlled fully automatically from the city’s control center based on traffic flow and time of day. But even such high-tech installations can be further improved through ultramodern control technology.

Intuitive Decision-Making. A far more advanced solution is a cognitive software system developed by Dr. Georg von Wichert, an automation expert at Siemens Corporate Technology. von Wichert loaded the system with four weeks’ worth of traffic data from Berlin. “In this case, ‘cognitive’ means that the system itself creates a model of the city’s traffic processes and then makes decisions,” he explains. In other words, the system does not base its assessment of the traffic situation on individual sensor measurements and on what’s happening on individual streets. Instead, it evaluates sensor data within the overall urban context and “understands” the situation as a whole. This is a form of intelligence that people use intuitively, which explains why service personnel like to switch back and forth between programs at the control center to get an overview.

The advantage of a cognitive system is that it can observe complex data in parallel and thus recognize deviations from normal congestion patterns faster than a human can. However, it must be trained. After von Wichert’s system had completed its learning phase, it became clear that the city’s traffic flows and regularly-occurring congestion could be accurately predicted. Special cases, such as jams resulting from accidents or short-term construction sites, were also reliably detected. Using this data, it would be possible to create an assistance system as an initial step, and thus to help control center personnel choose the best control program for the city’s traffic lights.

An additional advantage of the cognitive system lies the fact that it’s capable of learning. “In the next step, we could generate tiny variations in the system’s parameters,” says von Wichert. “This would allow us to test the reaction of the system as a whole and thus optimize the control programs in many small steps without interfering with traffic flow.”

More extensive tests would be able to show to what extent the traffic in a major city could flow more smoothly with the help of cognitive systems. But von Wichert is convinced that in complex situations, his learning model-based control system will be superior to human operators in a control center.

Diekmann also believes that automatic control systems will do far more than just manage road traffic. He predicts that in coming decades control centers for road traffic, local public transportation systems, and even the dispatch centers of courier and freight services will all be interconnected. “When that happens, cities will have a nervous system that makes it possible to comprehensively optimize traffic over broad areas,” he predicts.

If cognitive intelligence is one day used to coordinate all modes of transport and all transport hubs, the dream of the super mobility app might become a reality. This is how it would work: The traveler enters any destination in the world into the app, and the app suggests, for example, three different travel routes that can include any mode of transport, but in all cases are oriented toward optimized travel times, costs and CO₂ emissions. Such suggestions would be based not on theoretical schedules, but on current forecasts of traffic conditions. If conditions change in the course of a trip, route suggestions would be adjusted in real time.