# br Table br Patient mix per day br Patient type

2019-10-08

Table 8
Patient mix per day.
Patient type
Day
The total number of patients per time slot was calculated for both schedules (actual and proposed). This number includes both new and recurring patients at any point in time. The aim was to visualize the differences between the original schedule and the proposed schedule. Fig. 4 compiles the total number of patients for both schedules at any 20-minutes time slot. The red line represents the actual schedule, and the purple line represents the proposed schedule. From the figure, it BMS 986120 can be seen that the schedule obtained 
from the mathematical model (proposed schedule) is leveled when compared to the original.
The proposed schedule keeps a balance in the number of pa-tients in the system throughout the day and avoids peaks of de-mand. The proposed schedule was obtained for each one of the 10 days, and this leveling pattern manifested in 8 out of the ten days.
Given the success of the mathematical model to level patient demand throughout the day, it was decided to estimate the maxi-mum capacity for the system for each patient mix. Instead of using the number of patients per type as an input, the model used the percentage for each type of patient which allows for the allocation of the maximum number possible of each patient type according to the patient mix. Constraint (8) was then modified to achieve this objective. The new constraint became:
a
c
k
b
n 1 xijtan
n=1 yijtan
Where pj represents the percentage of each patient type according to the patient mix of the day. Table 9 summarizes the patient type percentages per day.
Table 10 presents the results of the comparison between the actual system and the maximum capacity that the infusion area could have managed on each of those days without exceeding the
Table 9
Patient percentage per day.
Patient type
Day
Fig. 4. Total number of patients per time slot day 1.
mental workload levels for nurses. From the table, Chromocenter can be seen that the mathematical model allows the capacity of the infusion area to increase by at least 50% per day
Additionally, the results show a variability among the days which is caused by the differences in the patient mix. Days 5 and 9 exhibited the largest percentage increase of capacity with values of 216.7% and 136.8% respectively. In contrast, days 1 and 7 do not show high percentages but they do exhibit increased number of patients per day with a value of 63 for both of them. Finally, day 3 showed a potential increase of only 50%. This is due to the fact that day 3 had many long-duration patients (four level 8, eight level 9, and two level 10).
4. Conclusions
This research study developed an optimization model for the infusion area in a cancer clinic and established general assump-tions and restrictions to design scheduling policies for patient appointments. A linear model was developed to obtain optimal schedules for the system.
In order to limit the level of mental workload for nurses in the current setting to no higher than 81.2 percentile of the observed instances, it is necessary to assign up to three patients per nurse at any given time. This value was part of the capacity constraints for the mathematical model.
The results of the mathematical model were used to generate scheduling policies based on the worst and the best scenarios. This model allows the cancer center to allocate patient’s appointments in a more balanced manner resulting in an increased capacity without significantly impacting the total average time per patient. Given the generality of this model, it can be used on similar cancer centers. The model has the capability to change the numbers of
Patient percentage per day (actual capacity (Cact), maximum capacity (CMax)).
Day Capacity