Effective employee performance evaluation is critical for organizations as it boosts productivity, assists in making informed HR decisions, and ensures fair appraisals. Traditional methods of performance assessment are generally subjective, time-consuming, and prone to errors. This research addresses these challenges by applying machine learning models to automate the prediction of employee performance and generate appraisal recommendations in a factory environment. The dataset was picked from five stations: Design, Machining, Forging, Pickling, Inspection, and Packing, and covers three periods of cycle for each employee. Four models, Gradient Boosting (GB), Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Network (ANN), were used in the study. These models were trained and tested on preprocessed data to classify employees as high and low performers, as well as predict appraisal scores accordingly. The results show that the accuracy of Gradient Boosting reached 97.86% while that of SVM stood at 94.5%, and for RF was 87.46% and for ANN, at 82.23%. Besides accuracy, efficiency varied between the models: while Gradient Boosting proved superior to the rest both in terms of latency and throughput, it is thus suitable for a real-time application for predicting the employee's performance. The findings underscore the potential of machine learning to enhance HR management practices through the automation of performance assessment, thereby improving decision-making, fairness, and the optimization of HR processes towards the overall improvement of organizational performance.