Enhancing Workforce Investment Through Deep Learning and Predictive Analytics in Workday Human Capital Management

In the current competitive business environment, organizations face significant challenges in effectively managing and investing in their human capital. Traditional human resource management models often fail to provide timely insights for employee performance, skill development, and retention, leading to suboptimal workforce utilization and increased turnover. To address these challenges , this study proposes an artificial intelligence (AI) based model that leverages deep learning and predictive analytics within workday human capital management (HCM) systems. The proposed work used to optimize workforce investment by accurately predicting employee potential, identifying skill gaps, and enhancing decision-making in recruitment, training, and retention strategies. The proposed model integrates a deep graph neural network (DGNN) model with an enhanced lotus effect optimization (ELEO) algorithm for feature selection and hyperparameter tuning, ensuring robust and efficient learning from complex organizational datasets. The Big-Five Personality Traits dataset is used to validate the model and preprocessing techniques are employed to ensure data quality and consistency. The proposed ELEO+DGNN model further enhances performance, achieving a mean testing accuracy of 99.620%, which corresponds to a gain of 0.370% over DGNN, 1.200% over SVM, 3.860% over RF, and 6.560% over KNN. Precision, recall, and F1-score also reach 99.620%, while training accuracy is 99.906%, indicating strong generalization and minimal overfitting. The comparative analysis confirms the ELEO+DGNN model as a robust and scalable solution for intelligent workforce management, providing actionable insights for HR professionals and organizational leaders.