In the data-rich but knowledge-poor domain of production management systems, the utilization of machine learning (ML) for lead-time prediction has gained increasing attention. Despite several efforts focusing on ML and regression techniques, the selection of features for lead-time prediction remains a challenge. The purpose of this study is to explore the socio-technical challenges and benefits of applying ML to predict lead-time in manually executed tasks in the biopharmaceutical industry, with a particular emphasis on the quality control of raw materials and semi-finished products. Through a case study and empirical analysis, the research identifies critical factors affecting lead-time prediction in manual tasks and evaluates the socio-technical implications of implementing ML-based solutions. Moreover, the study provides valuable insights into the practical challenges and potential advantages of adopting ML techniques for lead-time prediction in the biopharmaceutical sector, offering a comprehensive understanding of the complex interplay between technology and human factors. Finally, we discuss the implications of the findings for managers and staff responsible for the planning of manual tasks, providing actionable recommendations to improve production efficiency and lead-time prediction accuracy. This research contributes to the growing body of knowledge on the integration of ML in production management systems and highlights the need for further investigation to harness the full potential of ML in addressing the unique challenges of the biopharmaceutical industry.