A STOCHASTIC OPTIMIZATION MODEL TO IMPROVE PRODUCTION PLANNING AND RESEARCH-AND-DEVELOPMENT RESOURCE-ALLOCATION IN BIOPHARMACEUTICAL PRODUCTION PROCESSES

The increasing cost of health care has brought pressure to reduce phar maceutical costs, and because manufacturing and R&D are significant co st factors, these areas have been targeted as potential sources of cos t reduction. Manufacturing costs are particularly high in the biotechn ology industry because process technologies are relatively new. Contam ination, genetic instability, and other factors complicate production planning and make bioprocess systems unreliable. This paper presents a Markov decision process model that combines features of engineering d esign models and aggregate production planning models to obtain a hybr id model that links biological and engineering parameters to optimize operations performance. Using tissue plasminogen activator as a specif ic example, the paper shows how the hybrid modeling approach not only improves production planning, but also provides accurate information o n the operating performance of bioprocesses that can be used to predic t the financial impact of process changes. Therefore, the model can be used to guide investments in manufacturing process improvement and R& D (e.g., genetic modifications). Although stochastic production models are not commonly used in process design, this paper shows how a combi ned engineering production model can facilitate a concurrent design ap proach to reduce cost in bioprocess development.