Robust Advanced Modeling and Scheduling System (RAMS): From Space Exploration to Real-World Biomanufacturing

Abstract: Advanced planning and scheduling (APS) systems aim at helping production and operation managers in organizing the manufacturing process of biotech products, finding near-optimal schedules meeting the demand, while taking operational resources and raw materials into account. We introduce an extension to APS, called Robust Advanced Modeling and Scheduling (RAMS), and present the first RAMS system: Rombio. Unlike existing APS systems, Rombio allows to (i) visually model the operational problem and context entirely (ii) generate and optimize schedules while taking uncertainty into account, and (iii) deal with a combination of various key performance indicators (KPIs). Probability theory enables us to cope with uncertainty, computing schedules that are robust to temporal deviations. Depending on the pursued KPIs, the resulting schedules optimize a combination of the following terms: the probability of satisfying the process constraints, the expected return/efficiency/quality, and even the operators’ wellness by minimizing its expected extra-hours. This introduces a new risk-aversion paradigm, replacing the well-known what-if and sensitivity analysis frameworks. Initially developed in collaboration with Nasa for space exploration, this versatile tool is applied to real-world biotechnology manufacturing in three different contexts: diagnostics, medicines and stem cells.