Optimizing IEP Scheduling using Integer Linear Programming

Abstract

We address a common scheduling problem that special education teachers face: creating individualized education plan schedules (IEPs) for multiple students. The work began as a special project in an Optimization Modeling class and continued as an independent research project. The purpose was to develop a scheduling model that supports one special education teacher that works with thirteen fifth and sixth grade students. The model can be adapted for additional teachers or grade levels as needed. Currently, most IEP scheduling is a manual, iterative process that requires significant time and trial and error. This project aims to reduce the workload of special education teachers in creating IEP schedules through the usage of computational optimization. The model uses mixed integer linear programming implemented in AMPL, using the Gurobi branch-and-bound solver. The model defines sets of students, subjects, days, grade levels, and time blocks, and incorporates constraints including: preventing student double-booking, preserving teacher workload boundaries, supporting subject-specific instructional formats, and differentiating core subject priorities such as reading and mathematics. The objective function then balances these factors to produce a workable, conflict-free schedule. The resulting schedules meet IEP requirements, limit conflicts, and make better use of classroom time.