Rotation Order Matching Algorithm

This algorithm is designed to help third year medical students assign to the rotation order of their preference using a linear sum optimizer. The algorithm takes in the preferences of the students and the availability of the rotations, and assigns the students to the rotations in a way that maximizes the overall satisfaction of the students.

Requirements

In order to use this algorithm, you will need to have the following packages installed:

• Python3
• Pandas (for data manipulation and analysis)
• SciPy (for optimization functions)
• NumPy (for numerical computing)

You can install these packages by running the following command in the downloaded path:

pip3 install -r requirements.txt

Usage

To use the algorithm, the path of the preference matrix should be specified as the filename in the command line when running NRMA.py (see example below). The preference matrix should be a csv file with rows representing the students and columns representing the rotation preferences. The values in the matrix should represent the student's preference for the rotation, with higher values indicating a stronger preference.

The optimal rotation order assignment can be found by running the following command:

python3 NRMA.py batch_test.csv

Beans vs Linear Mode

The linear mode requires a preference.csv file with ranked preferences, while the beans mode requires a performance,csv file with assigned beans. More information about beans assignment can be found here.

Simulation File

Advanced testing for different penalty functions can be done via NRMA_simulate.py.

Example

Running python3 NRMA.py on the provided test file assigns a group of 8 students to 4 rotation orders. The output of this code should be:

  studentID  ...            rotation_order
0      abc6  ...  TBC1 – TBC3 – TBC2 – LAB
1      abc8  ...  TBC2 – LAB – TBC1 – TBC3
2      abc3  ...  TBC1 – TBC3 – TBC2 – LAB
3      abc4  ...  TBC3 – TBC1 – LAB – TBC2
4      abc1  ...  LAB – TBC2 – TBC3 – TBC1
5      abc5  ...  TBC3 – TBC1 – LAB – TBC2
6      abc2  ...  LAB – TBC2 – TBC3 – TBC1
7      abc7  ...  TBC2 – LAB – TBC1 – TBC3

[8 rows x 7 columns]
Average error of assignment for first rotation: 0.421875
Percent of students who recieved their first choice rotation: 0.625

Note that the order of students is stochastic.