Capacity planning under demand and manufacturing uncertainty for biologics

Abstract

Due to the long lead times and complexity in drug development and approval processes, pharmaceutical companies use long range planning to plan their production for the next 10 years. Capacity planning is largely driven by the long-term demand and its forecast uncertainty. The impact of uncertainties at manufacturing level, such as factory productivity and production success rate, are not entirely taken into account since only the average values of each manufacturing parameter are used. Can we better allocate production among manufacturing facilities when both demand and manufacturing uncertainties are considered? In this thesis a stochastic optimization approach is followed to minimize the deviation from target capacity limit under different manufacturing and demand scenarios. The mixed integer linear model incorporates the impact of demand and manufacturing variation on production allocation among manufacturing facilities through Monte Carlo generated scenarios. The thesis model is designed in a way that can be used as a decision tool to perform robust capacity planning at the strategic level.