Quality risk in the domain of drug production is a key factor contributing to patient safety and is critical for business success. Today, there is a lack of guidelines for its management at the industrial level. In recent years the limelight has been on the quality metrics draft guidance created by the United States Food and Drug Administration (FDA), as well as the industry-wide efforts on gathering, and analyzing operational data. Both trends have incited increased interest in the quantitative assessment of quality risk.
Existing guidance on Quality Risk Management in the pharmaceutical industry is typically focused on the micro-level of designing products and processes. On the other hand, Operations Management scholars provided research on the econometrical level. To the best of the author’s knowledge, there is no published research yet that examines site internal operational characteristics, and their relationship to quality risk in terms of quality compliance evaluations.
The main goal of this research is to develop a data-driven management approach for Quality Risk Management at the level of the manufacturing site. In order to achieve the research objective, four major steps were followed. First, a literature review of the known interlinks of inspection outcomes and characteristics of manufacturing operations was performed. Secondly, the relationship between Operational Excellence and quality risk is evaluated. Thirdly, the operational performance indicators are analyzed with regard to their possible link to inspection outcomes. And lastly, the applicability of predictive modeling in this Quality Risk Management context is tested and discussed.
This research project is designed to follow a mixed-method approach combining qualitative and quantitative research, as well as the solution prototyping phase. An iterative process enables establishing a first-hand understanding, preparing credible analysis results, and improving upon it to get a thorough understanding of the issue. Employing a systems theory view enables the assessment of quality risk in manufacturing sites as an analytically measured system’s state. Following the design science research methodology allows developing analytics methodologies, which will in turn facilitate and provide support to the decision-making process of the industrial managers.
The thesis closes with practical implications for managers who want to develop their quality risk management into a proactive and data based function.
Using Data Science to enhance Quality Risk Management in the Pharmaceutical Industry
Quality Risk Management in the pharmaceutical industry has a crucial responsibility for the safety of patients, as well as it is business-critical. Traditionally, it is often reactive and based on qualitative methods. This research enables proactive management and inspection readiness based on applying data science methods on operative performance indicators from production and QC labs. Companies and regulators can achieve higher drug safety standards while improving drug availability and costs.