An Integer Linear Program for Periodic Scheduling in Universities

Sina Moradi

2024-12-17

Abstract:Efficient scheduling of periodic meetings is a critical challenge in various service-oriented domains, including academic settings, healthcare, and legal consultancy. This study presents a robust Integer Linear Programming (ILP) model to optimize the scheduling of faculty-student meetings. The proposed model incorporates practical constraints such as minimum intervals between consecutive meetings, differing time requirements for undergraduate, masters, and PhD students, and dedicated emergency time slots for unplanned visits. The objective function aims to achieve an equitable distribution of meetings throughout the planning period while prioritizing earlier time slots and seamlessly integrating emergency appointments. To validate the effectiveness of the model, both numerical examples and a case study are examined. The results highlight the ability of the model ability to generate optimal schedules within a computationally efficient framework, leveraging the power of Gurobi optimization software. The model demonstrates significant versatility, extending its applicability beyond academic settings to any scenario requiring periodic scheduling of services, such as patient visits in healthcare or client consultations in legal practices. Future extensions of the model may include dynamic scheduling to adapt to real-time changes in availability and the coordination of schedules across multiple service providers.

Discrete Mathematics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the efficient scheduling of regular meetings between teachers and students in universities. Specifically, the author proposes an integer linear programming (ILP) model to optimize the meeting arrangements between teachers and students in order to meet the following challenges:

1. **Evenly Distributed Meetings**: Ensure that meetings are evenly distributed throughout the planned period and avoid being concentrated on certain days or time periods.
2. **Meet Different Needs**: Reasonably arrange the meeting duration and frequency according to the needs of students at different levels (undergraduates, master's students, and doctoral students).
3. **Reserve Emergency Time Slots**: Reserve specific time periods for unplanned emergency meetings to deal with unexpected situations.

### Specific Problem Description

- **Minimum Interval Requirement**: Specify the minimum number of days between different student categories to avoid overly frequent meetings.
- **Different Time Requirements**:
- Undergraduates need \( R_x \) consecutive half - hour time slots for each meeting, and \( T_x \) meetings need to be arranged during the planning period, with at least \( B_x \) days between two meetings.
- Master's students need \( R_y \) consecutive half - hour time slots for each meeting, and \( T_y \) meetings need to be arranged during the planning period, with at least \( B_y \) days between two meetings.
- Doctoral students need \( R_z \) consecutive half - hour time slots for each meeting, and \( T_z \) meetings need to be arranged during the planning period, with at least \( B_z \) days between two meetings.
- **Emergency Time Slot Arrangement**: Reserve \( L \) time slots per day for emergency meetings, and these time slots are arranged after all planned meetings on the same day.

### Objective Function

The objective function of this model aims to achieve the following three goals:
1. **Evenly Distributed Meetings**: Ensure that meetings and emergency meetings are evenly distributed among all working days.
2. **Give Priority to Early - Time Arrangement**: Give priority to arranging meetings in the earlier time periods of a day.
3. **Reasonably Arrange Emergency Meetings**: Arrange emergency meetings in the earliest available time slot after regular meetings.

The formula of the objective function is as follows:

\[
\text{Minimize } 2 \cdot \left( \sum_{j = 1}^{D} \sum_{k = 1}^{P} k \cdot \left( \sum_{i = 1}^{N_x} X_{ijk} + \sum_{i = 1}^{N_y} Y_{ijk} + \sum_{i = 1}^{N_z} Z_{ijk} \right) \right) + \sum_{j = 1}^{D} \sum_{k = 1}^{P} k \cdot E_{jk}
\]

Through this model, the author hopes to strike a balance between computational efficiency and practical application, thereby providing a solution applicable to academic institutions and other fields that require regular scheduling services.