The Analytics & Optimization (A&O) research group is dedicated to the entire spectrum of decision-making processes, with a pivotal focus on optimization techniques. Our expertise is quite wide, spanning stochastic processes, optimization, simulation-based approaches, and machine-learning algorithms.
Our research is driven by practical challenges across diverse domains, positioning us at the forefront of both theoretical developments and applied solutions. Here are some key application areas:
- Healthcare: We have successfully improved ambulance deployment, hospital patient flow, and long-term care management by optimizing processes like appointment scheduling and bed management. Our partnerships with the VU Medical Center and PICA focus on reducing patient waiting times across elderly, mental health, and youth care. To face these challenges, we use and develop new models from queueing theory, stochastic optimization, and machine learning.
- Production and logistics: In many industries production and transport are nowadays closely intertwined and failing to integrate them in a coherent schedule can lead to excessive costs. We develop scalable, efficient models for integrated production and transport scheduling, aiming to reduce costs and CO2 emissions. In the maritime sector, our research supports the transition to autonomous, zero-emission ships by creating methods that minimize future modification costs for systems and ship layouts.
- Service industry: Services involve high variability, both in arrival times and service duration, making necessary the use of appropriate stochastic models for analyzing these customer and work pipelines. We specialize on all aspects of the planning process (forecasting, prediction, optimization) and on applications in different sectors (e.g., call centers, aviation, health providers). Over the years we have built strong research collaborations with partners in all these sectors.
- Energy networks: As energy networks are rapidly evolving due to ambitious sustainability goals, we address challenges posed by renewable energy fluctuations and extreme weather. Using stochastic optimization and reinforcement learning, we develop strategies for adaptive topology control and energy reserve sizing to maintain grid reliability.
- Other emerging data-driven applications: Our researchers are pioneering in areas influenced by the surge in data availability, such as social media analytics, public transportation systems, and human resource management. Projects include detecting Twitter/X anomalies, identifying shifts in workforce dynamics, and understanding group behavior in public transit systems.
We are committed to pioneering innovative solutions that not only advance academic understanding but also provide tangible benefits to society. Through our collaborations with various industry partners and academic institutions, we continue to lead in both the creation of new knowledge and the practical application of our research findings. A considerable part of our research is in collaboration with the Stochastics group at CWI.