Workforce scheduling in nursing is among the most demanding planning tasks in hospital operations. A valid roster must simultaneously satisfy a large number of (conflicting) requirements: statutory working-time and rest-period regulations, qualification- and ward-dependent minimum staffing levels, vacation and absence periods, fairness among team members, and individual preferences. In practice, this balancing act is still frequently performed manually or with only limited software support.

The result is a process that is time-consuming and that heavily ties up the staff responsible for planning, and that leaves considerable optimisation potential untapped. While the underlying problem is well known in operations research, the gap between formal models and tools that are usable under real organisational conditions remains substantial.

The goal of this thesis is to develop a prototype based on qualitative interviews and/or a literature review for the automated or semi-automated creation of nursing schedules under the constraints described above. The relevant requirements will first be elicited and structured into hard and soft constraints, and the scheduling task will be formalised as an optimisation or constraint-satisfaction problem. This thesis provides the opportunity to cooperate with a local hospital in Münster.

If you are interested, please send your current transcript of records together with a brief statement of motivation.