Throughout the entire project, relevant data is collected and anonymized from the partner organization to serve as a foundation for the residents’ planning process. To identify criteria for fairness and stability, as well as potential weaknesses, experts are interviewed, and the results are used to develop mathematical models. Additionally, interviews are conducted with residents to assess their satisfaction and gather feedback on the project. The insights gained from these interviews are used to identify and address the project’s strengths and weaknesses, thereby supporting the achievement of its intended goals.

In the first step, it is important to clearly define the demand for and capacity of surgical procedures, not only by shift but also with regard to the different types of physicians. This is necessary because the number of surgeries a resident can perform within a given time period and at a specific clinic is a fundamental component of completing the various stages of training. Therefore, realistic projections help in planning the duration of the various stages.

Key figures are an essential operational tool for comparison and regulation, as well as for measuring the achievement of targets. Therefore, efforts are made to derive key figures from other service areas, use them to model objectives, and evaluate interdependencies. In this context, thresholds should help evaluate the suitability of the key figures. In collaboration with the project partner, a method for tracking these key figures as part of daily operations is being developed.

In order to apply the project’s findings to hospitals, a suitable interface for workforce planning software is necessary. Therefore, the correlations between education and workforce planning are analyzed and mapped to an interdependency matrix. This also helps identify and avoid potential problems and is intended to simplify future research.
In addition, a guideline for researchers and practitioners summarizes the challenges arising in resident planning to help them understand the complexity of the planning process.

A clear educational structure and reliable planning enhance a hospital’s reputation for providing attractive and efficient training, thereby giving it a competitive advantage in recruiting physicians. However, hospitals must also ensure high levels of patient care and the effective use of human resources. Addressing all these aspects requires, as a first step, determining the optimal number of training positions while taking resource and capacity constraints into account. Due to the complexity of the planning process, mathematical models and methods are used to first determine the optimal number of resident positions along with the appropriate training schedules. Therefore, a mixed-integer problem is formulated and solved using OR/MS algorithms, such as column generation, branch-and-price-and-cut, or Bender’s decomposition. The developed methods are tested through simulations as well as case studies at the partner institution.

After developing the mathematical model for resident scheduling, the problem is expanded to include stochastic factors that occur in reality and affect demand and capacity. Therefore, on the one hand, Markov decision processes are used, and on the other hand, a two-stage stochastic programming approach, which aims to provide more ex ante information. The previously developed methods need to be adapted to these issues.
However, uncertainties arise not only at the strategic planning level but also at the tactical planning level, where the educational schedule is transformed into a concrete work roster. In this context, measures of certainty help to avoid unfairness in terms of steady progress or the number of clinic transitions.
Overall, the uncertainties regarding the number of surgeries are intended to be addressed through a robustness concept in order to improve planning reliability and reduce the number of roster changes.

Since fairness directly affects employee satisfaction and, consequently, the quality of work, various concepts of fairness are defined and their impact on personnel planning is evaluated. In this context, particular attention is given to the number of clinic transfers and the intended impact on the continuity of care. Therefore, the developed models are adapted and solved heuristically, for example using local search heuristics.