Stochastic methods of operations research represent the focus or our research.
Accordingly, we develop models describing technical or economic problems that are exposed to random (or unknown) influences. Under such circumstances stochastic modeling becomes necessary since deterministic models fail to capture random effects. The developed models provide important characteristics of the underlying problem and an optimal strategy can be deducted.
Our research activities currently focus on the following areas of application an associated problems:
-
Airline Revenue Management
- Passenger and Cargo Capacity Control
- Dynamic Pricing
- Demand Driven Dispatch
-
Health Care
- Patient Scheduling
- Queuing Models
- Medical Test Design
- Clinical Decision Making
In order to model and solve such complex problems we mainly apply the following methods:
- Markov Decision Processes (MDP)
- MDPs with non-standard optimality criteria
- Partially-observable and Semi-MDPs
- Approximate dynamic programming
- Reinforcement learning
- Optimality of simply structured decision functions
You can find further information on each researcher's website.