A Data-Driven Framework FOR Optimizing It Absorptive Capacity IN Schools: Integrating Neural Networks WITH Genetic Algorithms

Abstract

This paper develops a theoretical model of a combined method based on machine learning and optimization to improve IT absorptive capacity (ITAC) and allocate resources using the Iraqi schools’ dataset. Strategically important indicators of the network bandwidth, numbers of devices, system availability, budgetary provisions and the IT literacy of staff are also part of the dataset which serve as the basis for precise demand calculation and maximization. Our framework incorporates an IT resource forecasting neural network (NN) that has low mean squared error (MSE) and high R2, and a genetic algorithm (GA) for efficient distribution of these resources. The GA was also successful in cutting overall operating costs by an average of 18% while raising ITAC by only 25 % to illustrate its success in low-resource educational environments. Budget allocation and bandwidth utilization were sources of variation in absorptive capacity as established through the sensitivity analysis: These enhancements provided tangible information for strategic investment priorities. This adaptive, data-driven approach offers not only a practical solution to a multitude of IT management issues in schools but also a feasible path toward building more robust and inexpensive infrastructure for education.