Previous research work in the area of capacity planning and scheduling of biopharmaceutical manufacture has been mostly based on discrete-time mixed-integer linear programming (MILP) formulations. This paper presents a multi-objective optimisation approach that utilises an alternative continuous-time based model of the bio-manufacturing process and a genetic algorithm (GA) to optimise medium-term production throughput and product inventory levels for a multi-product biopharmaceutical facility. The model accounts for constraints and features such as sequence-dependent product changeovers, multiple intermediate demand due dates based on a realistic monthly schedule, and product QC/QA release times. Using the proposed approach, detailed production schedules and Gantt charts can be produced with the exact dates assigned to the important stages of any given production campaign such as the start and end of the first batch, the first harvest, and the estimated QC/QA release of a batch. An industrial case study developed with Eli Lilly and Company is used to illustrate the functionality of the model. The case study focuses on capacity planning taking into consideration different stainless steel and single-use technology capabilities. The model is compared with the traditional spreadsheet-based planning tool, which it outperforms in terms of improved throughput, product inventory levels and solution time. The proposed model is also used to consider the potential of supply or demand uncertainties present in a typical bio-production process.

Multi-objective medium-term capacity planning for stainless steel and single-use multi-product biopharmaceutical facilities under uncertainty using a genetic algorithm

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.