Genetic algorithm optimizes high school math resource allocation under time constraints
New GA solves NP-hard problem of assigning personalized materials to students within study time limits.
The paper, part of the Erasmus+ KA220-SCH project, addresses the challenge of allocating educational resources to individual students to accommodate learner diversity. The authors formalize the selection of mathematics materials for high school students as a 0-1 binary combinatorial optimization problem, constrained by strict study time limits. This formulation is NP-hard, meaning exact solutions become computationally infeasible as the catalog of resources grows. To solve it practically, the team proposes a Genetic Algorithm (GA) that incorporates a custom constraint repair mechanism. The GA efficiently explores the binary decision space, ensuring allocated resources respect time constraints.
Experimental evaluation across 10 independent runs demonstrates the algorithm's fast convergence to high-quality solutions, with strong stability across different random seeds. The results confirm the viability of metaheuristics for real-time educational decision-support systems. This work provides a practical tool for schools to dynamically tailor resource sets to each student's needs, potentially improving equity and learning outcomes without overwhelming administrators or teachers.
- Models resource selection as a 0-1 binary combinatorial optimization problem with study time constraints.
- Uses a Genetic Algorithm with a specialized constraint repair mechanism to solve the NP-hard problem.
- Achieves fast convergence and high solution quality across 10 independent experimental runs.
Why It Matters
Enables real-time, personalized resource allocation for diverse learners, potentially improving equity and efficiency in secondary education.