Transportation of people and goods is becoming increasingly important in today's interconnected world.
The way transportation systems are organized has a large impact on the economy, the environment, and the way of life of a given city, community and every individual.
We are contributing to IBM's Smarter Transportation agenda by solving real-world planning and control problems in transportation and logistics using our deep skills in mathematical modeling and optimization techniques. With profound industry knowledge about how transportation systems function and how they can be designed, controlled and coordinated, we tackle the challenge of how to cope with the complex dynamics in the context of various constraints, such as limited time, incomplete state information, stochastic disturbances, or merely the local influence of control actions.
Railway scheduling
In many countries, railway traffic for both passenger and freight transportation has been increasing considerably, and this trend is expected to continue. As building new infrastructure is very expensive and hardly possible in many urban areas, the capacity of the existing network must be better utilized to meet the growing customer demand for more transportation services.
In various projects, we work on advancing the state-of-the-art for planning and control of railway operations, for instance for the automatic construction and optimization of timetables as well as automatic control methods for dynamic scheduling to be used as decision support for dispatchers in control centers.
References
- G. Caimi, M. Fuchsberger, M. Laumanns, M. Lüthi: A model predictive control approach for discrete-time rescheduling in complex central railway station areas. Computers & Operations Research 39(11) 2578-2593, 2012. DOI
Passenger routing
Challenge
Traditional routing approaches assume that the exact behavior of the underlying transportation network is known, whereas in reality, the level of detail at which such a system is managed ranges from completely deterministic to almost random. In the latter case, the stochasticity is due to many factors such as random breakdowns, unknown traffic behavior, or unknown starting times of transportation vehicles. This results in stochastic input parameters such as random travel times between stops or random arrival times of transportation vehicles.
Solution
IBM Research has designed and implemented several approaches to deal with different aspects of stochastic transportation networks. The computed routing policies show a clear advantage over routing policies that are computed by traditional approaches, which do not take stochastic behavior into account.
Benefits
Allows the user of a stochastic transportation network to reach a destination in a shorter amount of time, but also in a more convenient manner with respect to other quality factors.
