Sistem Pemantauan Cerdas untuk Evaluasi Kinerja Karyawan: Analisis Prediktif Berbasis Deep Learning

Abstract

Employee performance evaluation is a critical process within organizations; however, the traditional methods still in use tend to be subjective and fail to make optimal use of historical data. This study aims to design and develop a Deep Learning-based intelligent monitoring system capable of supporting employee performance evaluations in a more objective, adaptive, and data-driven manner. The research method used is Design Science Research (DSR), which includes the stages of problem identification, goal setting, design and development, demonstration, and system evaluation. The system was developed by integrating Angular frontend technology, .NET Core Web API backend, SQL Server database, and a Deep Learning model based on Long Short-Term Memory (LSTM). The system is capable of generating several key outputs, namely efficiency scores, employee performance categories, and predictions of future workloads based on historical data patterns. System evaluation using the System Usability Scale (SUS) method yielded a score of 74, indicating that the system has a good level of usability and is acceptable to users. Thus, the developed system is capable of serving as a more objective performance evaluation solution and supporting data-driven decision-making.