Development of Decision Support System on Forecasting Workload Analysis using Time-series Algorithm

Abstract

The growing dynamics of faculty workload planning at Isabela State University (ISU) must be addressed, this study provides the development and implementation of a Decision Support System (DSS) with time-series forecasting to enable better academic resource planning. With ISU’s dynamic fluctuations in student enrollment and diversified programs, manual methods have failed to cope with dynamic workload allocations among faculty members of ISU. This study aims to bridge that gap by integrating historical information with forecasting analytics through an ARIMA-LSTM hybrid model using different metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The DSS analyzes trends in class enrollments, faculty assignments, and enrollment records to predict upcoming faculty workloads. RMSE, MAE, and MAPE metrics confirm the high accuracy of model forecasts with MAPE as low as 1.25%. The DSS was predictive accuracy-tested and usability-tested, offering a more data-driven insights for university administrators to make more proactive and informed decisions within ISU system