Batch Seat Swapping

The batch seat swapping algorithm applies to students who have been assigned a seat in the CSE RPg Labs and voluntarily participate in the scheme. The algorithm searches for improved seat arrangements among participating students so that those who share the same primary supervisor move closer.

Step-by-Step Instructions

Step 1: Prepare Input Files

Input files should be placed under the input/ directory using the following file names:

• File input/students_seats.json: List of student-seat pairs assigned by the initial RPG Lab seat application process.
• File input/seats.json: List of all RPG Lab seats that are available to CSE. These seats contain both occupied ones (i.e., those that have been assigned in input/students_seats.json) and empty seats (i.e., those that have not been assigned to any student).
• File input/students_fac.json: List of student-faculty pairs. For students that are co-supervised by multiple faculty members, their corresponding faculty members in this file are their primary supervisors.
• File input/participate.txt: List of students who sign up for batch swapping. (Students who sign up but don't currently have an assigned seat will be disregarded in the swapping.)

We have included a set of sample input files for illustration and testing purposes. Please replace them with authentic data.

Step 2: Run the Batch-Swapping Script

python3 assign.py

Step 3: Export the Output Files

The most important output file is this one:

• File output/swapped.json: List of updated student-seat pairs after batch swapping.

Please load this file into CSSystem.

Please disregard other files under the output/ directory, as they are generated for debugging and illustration purposes only:

• File output/optimized.html: Visual representation of updated seat-student pairs.
• File output/optimized.json: List of updated seat-student pairs.
• File output/initial.html: Visual representation of initial seat-student pairs.
• File output/initial.json: List of initial seat-student pairs.

Additional Information

Please note that while the scheme aims to reduce the average within-group distance for all participating students, there are no guarantees for individual success, fairness, or precise seat proximity calculations. In particular, the current limitations are as follows:

1. There are no guarantees on individual success rates; some students may not experience improved seat proximity to their group-mates.
2. While participating, some students may be initially assigned a desirable seat (e.g., next to a window) but later be swapped to a less desirable seat (e.g., next to a hallway). The newly assigned seats may be in different rooms than those initially assigned. If a student participates, they must move to the newly assigned seat.
3. Seat proximity is currently calculated by subtracting room and seat numbers, which is less accurate than measuring physical distance. Due to limited manpower for this exercise, we are unable to provide precise calculations this year but hope to improve this aspect in the future.

Current algorithm: Random hill climbing.