I work in Sequential Decision Analytics, focusing on problems with high-dimensional, discrete decision spaces from Operations Research. To solve those problems, I apply Stochastic Programming and Reinforcement Learning. I apply my knowledge to Healthcare Logistics problems such as capacity allocation during infectious outbreaks and scheduling of surgeries or appointments.

Expertise

  • Nursing and Health Professions

    • Time
    • Patient
    • Hospital
  • Computer Science

    • Heuristics
    • Linear Program
    • Programs
    • Service
    • queueing model

Organisations

Publications

2022

Surgical case mixes and distributions of perioperative surgical process durations for German hospitals (2022)[Dataset Types › Dataset]. Zenodo. Korzhenevich, G., Zander, A. & Bialas, E.https://doi.org/10.5281/zenodo.7147921Zeitaufwand des Impfprozesses und der Organisation der COVID-19-Impfung in Hausarztpraxen (2022)Zeitschrift fur Allgemeinmedizin, 98(3), 100-105. Buhlinger-Göpfarth, N., Zander, A., Heckmann, I., Holzmann, T., Lauck, K., Stengel, S., Nickel, S. & Peters-Klimm, F.https://doi.org/10.53180/zfa.2022.0100-0105

2021

OP-Planung – der Einsatz quantitativer Methoden zur Entscheidungsunterstützung (2021)OP-Management up2date, 1(04), 327-339. Korzehnevich, G., Zander, A., Nickel, S. & Schuster, M.https://doi.org/10.1055/a-1639-3848Managing the intake of new patients into a physician panel over time (2021)European journal of operational research, 294(1), 391-403. Zander, A., Nickel, S. & Vanberkel, P.https://doi.org/10.1016/j.ejor.2021.01.035Demand and Capacity Management for Medical Practices (2021)[Thesis › PhD Thesis - Research external, graduation external]. Karlsruhe Institute of Technology. Zander, A.https://doi.org/10.5445/IR/1000137945

Research profiles

I am responsible for the Master course on Reinforcement Learning, which focuses on the ODE approach of Stochastic Approximation and how it can be applied to show the convergence of Reinforcement Learning algorithms. Further, I co-organize the Bachelor course Topics in Sequential Decision-Making.

In addition, I supervise Bachelor, Master and Internship students, mostly working on topics within Healthcare Logistics.

Affiliated study programs

Courses academic year 2024/2025

Courses in the current academic year are added at the moment they are finalised in the Osiris system. Therefore it is possible that the list is not yet complete for the whole academic year.

Courses academic year 2023/2024

Current projects

Regular and Unplanned Care Adaptive Dashboard for Cross-Border Emergencies

During cross-border health emergencies, health and care services may be overwhelmed by high numbers of patients requiring unplanned care. Delays and backlogs in regular care, as a result of the stretched healthcare system, leave millions of patients with regular care needs unattended, resulting in disastrous healthcare outcomes. If poor healthcare outcomes across the population are to be avoided, healthcare systems must become more resilient and flexible and allow for rapid changes in the care delivery services. RAPIDE aims to develop, validate, and demonstrate a portfolio of powerful tools that enable healthcare systems to increase the robustness of decisions, the resilience of healthcare professionals and patients, and the flexibility in the modalities of care delivery, thereby maintaining access to regular care during health emergencies. RAPIDE emphasizes opportunities for optimizing in-hospital care but also for relocating care from hospitals to community and home environments without loss of care quality. Thus, the project focuses on two closely linked challenges – 1: Identifying and predicting how much care and which care needs to be moved along the care chain; 2: Identifying and verifying effective, feasible, and acceptable ways to make this reconfiguration of care a reality. This will be achieved by (a) resource modeling, which builds comprehensive foresight and forecasting solutions and links them to patient flow optimization along the whole care chain, and (b) selecting and implementing the best available tools to deliver regular care in new ways. RAPIDE will be co-designed and co-validated with stakeholders, from patients, GPs, clinicians, and hospital managers to health ministries, pandemic management, and public health agencies, to ensure usability, acceptability, and equitable real-world value.

Supporting Efficient Deployment of Nursing Home Staff Through Demand Prediction

Hospitals are discharging many patients that are in need of aftercare, e.g., requiring a bed in a nursing home. The timely transfer of those patients to the correct type of aftercare is of utmost importance to ensure the quality of care, i.e., that the patient receives the right treatment at the right time, as well as the quality of staffing, i.e., that the capacity and staff in the hospitals and the aftercare organizations are used efficiently. If patients do not receive timely treatment, it may negatively impact their recovery journey. Further, they stay in the hospital bed longer than necessary, producing so-called bed-blocking days. These blocked beds are not available to other patients in need, and highly trained and costly staff needs to take care of those patients instead of performing tasks corresponding to their skill set. This project focuses on developing models to predict demand for nursing home care originating from hospitals on different time scales (long-term, mid-term, and short-term). Those hospital predictions will be made available to nearby nursing homes in order to support their capacity and staff planning. By accurately predicting demand and aligning resources accordingly, the project seeks to reduce bed-blocking, improve patient outcomes, and enhance the efficiency of staff planning in nursing homes and hospitals. The project involves collaboration with multiple hospitals to ensure its applicability and aims to develop a prototype prediction tool for immediate use, with the long-term goal of integrating the models into hospital information systems for widespread implementation.

Improving regional patient admissions and inter-regional patient transfers during a pandemic

Improving inter-regional patient transfers during a pandemic represents a comprehensive and forward-thinking endeavor aimed at elevating the efficiency and equity of inter-regional patient transfers during pandemics. Operating within the broader context of crisis response and infectious disease outbreaks, particularly pandemics, the project is poised to make substantial advancements by developing prediction models for both infectious and non-infectious hospital arrivals and bed censuses. Complementing these predictive capabilities is the creation of sophisticated optimization models specifically designed to guide decision-making for interregional transfers when care resources are in high demand, with a central objective of ensuring an impartial and well-distributed allocation of care across diverse regions also taking into account non-infectious care. At its core, the project is not merely a technical endeavor; it carries profound societal implications by emphasizing the fundamental right to equitable care access during times of crisis, asserting the significance of care as a human right that must be upheld even in the face of pandemics. On the scientific front, the project's contributions extend beyond the routine, as it delves into the refinement of short-term prediction methodologies and introduces innovative optimization models, all underpinned by a collaborative, interdisciplinary approach. The synthesis of expertise from diverse fields, such as infectious diseases, mathematics, and operations management, reflects a strategic and holistic response to the multifaceted challenges presented by pandemics. The model approach weaves together statistical methods, queueing theory, stochastic optimization models, and simulation techniques. The project's feasibility and applicability are orchestrated through strategic partnerships with key stakeholders, including hospitals, ROAZ (Regionaal Overleg Acute Zorgketen, Regional Consultation Body on Acute Care) regions, and LCPS (Landelijk Coördinatiecentrum Patiënten Spreiding, National Patient Spread Coordination Center)/LNAZ (Landelijk Netwerk Acute Zorg, National Acute Care Network). The implementation of a cloud-based solution for data control stands out as a testament to the project's commitment to safeguarding patient privacy, while the integration of a decision support system within LCPS enhances its potential for widespread national application. The partners' collective experience forms a solid backbone, instilling confidence in adhering to the project timeline, while proactive knowledge transfer activities are poised to amplify the dissemination and retention of invaluable project outcomes.

In conclusion, the project presents a holistic approach for interregional patient transfer in crisis times, leveraging data-driven models, interdisciplinary collaboration, prototype implementation, and stakeholder engagement to enhance patient allocation during infectious outbreaks, with implications for future research and care delivery beyond the project duration.

Strategic research initiative on real-world-inspired sequential decision-making

The strategic research initiative within 4TU.AMI focuses on establishing a community of researchers with different mathematical backgrounds to advance the development of new solution methods for sequential decision-making problems and prove performance guarantees of those methods, with a special focus on problems that exhibit real-world characteristics such as those found in finance, agriculture, and healthcare. The initiative seeks to address gaps in research and collaboration. It emphasizes methodological development with regard to uncertainty, large-scale systems, and non-stationarity, e.g., due to multiple agents, aiming to bridge theoretical advancements and practical implementations.

Scan the QR code or
Download vCard