Digital Management of Educational Resource Allocation in Colleges and Universities Based on PSO Algorithm

Han Yu2024 · DOI

Abstract

The educational resources of colleges and universities (CU) are an important manifestation of the country's comprehensive strength. In order to make all aspects of the school better serve teaching, provide the people with high-quality, efficient, high-quality and high-level learning environment, and optimize the allocation of these resources, this paper proposes a digital management framework based on the Particle Swarm Optimization (PSO) algorithm. By modeling resource distribution as a dynamic optimization problem, the system maintains stable performance—keeping database CPU utilization below 20%—even under significant concurrent user loads.

Introduction

Higher education institutions currently face a significant "resource gap." While the demand for high-quality, digital-first learning environments is skyrocketing, the infrastructure to manage these resources often lags behind. In many universities, information remains siloed across departments, leading to inefficient distribution of teachers, equipment, and digital content. Because resource allocation is a dynamic process influenced by shifting student demographics and limited capital, traditional rigid management systems often result in wasted resources. This research bridges this gap by applying swarm intelligence to create a scientific, automated allocation model.

System Design and Functional Modules

The proposed digital management (DM) system is built on a Browser/Server (B/S) three-tier structure, which effectively separates the user interface, business logic, and data storage layers. This architecture ensures scalability and ease of maintenance.

The system's core functionality is divided into two primary modules:

  1. Resource Management Module: This component handles the lifecycle of educational resources, including auditing, publishing, storage management, and information classification.
  2. System Management Module: This component ensures the smooth operation of the platform by managing logs, databases, announcements, and general system maintenance.
Functional modules of the digital management system including resource and system management components.

Optimization via Particle Swarm Algorithm

The heart of the allocation logic is the Particle Swarm Optimization (PSO) algorithm. PSO is a computational method that optimizes a problem by iteratively improving candidate solutions based on a given measure of quality, mimicking the social behavior of bird flocking. In this system, educational resource "flow" is modeled similarly to data packets in a network, utilizing standard PSO velocity and position update equations to navigate the search space of resource configurations. This macro perspective allows the algorithm to coordinate disparate departments that would otherwise compete for the same budget or personnel.

Performance Evaluation and Results

The system was evaluated using experimental modeling in Matlab and stress-tested with QALOAD to simulate concurrent user environments ranging from 110 to 170 users.

The testing revealed that the application server, which handles the complex PSO logic, is the primary performance driver, consistently operating at 76-79% CPU utilization. In contrast, the database server remained highly efficient, with utilization staying below 20%.

System performance metrics, including server utilization and error rates, under varying concurrent user loads.

As shown in the data, even at the maximum test load of 170 concurrent users, the system error rate remained low at 4.5%. Furthermore, the study found that increasing initial system capacity significantly improves efficiency; for instance, increasing initial capacity from 30 to 50 dropped main program CPU utilization from 30% to just 6% when response times exceeded 9 seconds.

Conclusion and Future Work

The PSO-based digital management system provides a stable and efficient framework for university resource allocation. While the current results are promising, future work should focus on scaling the system to support thousands of simultaneous users and refining the PSO algorithm with genetic mutations to avoid local optima. Additionally, transitioning from simulated Matlab environments to live campus pilot studies will be crucial for validating the system against real-world administrative factors.

Summary updated

Abstract

The educational resources of colleges and universities(CU) are an important manifestation of the country's comprehensive strength. In order to make all aspects of the school better serve teaching, provide the people with high-quality, efficient, high-quality and high-level learning environment, and
BibTeX
@misc{yu2024digital,
  author = {Han Yu},
  title = {{Digital Management of Educational Resource Allocation in Colleges and Universities Based on PSO Algorithm}},
  year = {2024},
  doi = {10.4108/eai.17-11-2023.2342755},
}
Figure