In outpatient chemotherapy, nurses administer the drugs in two steps. In the frst few minutes of each appointment, a nurse prepares the patient for infusion (drug administration). During the remainder of the appointment, the patient is monitored by nurses and if needed taken care of. One nurse must be assigned to prepare the patient and set up the infusion device. However, a nurse who is not busy setting up may simultaneously monitor up to a certain number of patients who are already receiving infusion. The prescribed infusion durations are signifcantly diferent among the patients on a day at a clinic. We formulate this problem as a multi-criterion mixed integer program. The appointments should be scheduled with start times close to patients’ ready times, balanced workload among nurses, few nurse changes during appointments, and few nurse full-time equivalent (FTE) assigned to the schedule of the day. As the number of nurse FTEs is an output of the model rather than a fxed input, the clinic can use the nursing capacity more efciently, i.e., with less labor cost. We develop a 3-stage heuristic for fnding criterion points with the minimum weighted average deferring time of appointments for the minimum feasible number of nurse FTEs or a desired value above that. By not constraining the number of chairs or beds, we can fnd solutions with better (dominating) criterion points. Drug preparation, oncologist visit, and the laboratory test can be scheduled based on the drug administration appointment start time. Thus, the drug administration resources are efciently used with desirable performance in taking the interests and requirements of various stakeholders into consideration: patients, nurses, oncologists, pharmacy, and the clinic.
- Health services, Integer programming, Multi-criterion, Optimization, Scheduling, Chemotherapy
ASJC Scopus subject areas
- Management Science and Operations Research