Biopharmaceutical manufacturing is a rapidly growing industry with impact in virtually all branches of medicine. Biomanufacturing processes require close monitoring and control, in the presence of complex bioprocess dynamics with many interdependent factors, as well as extremely limited data due to the high cost of experiments and the novelty of personalized bio-drugs. We develop a new model-based reinforcement learning framework that can achieve human-level control in low-data environments. A dynamic Bayesian network is used to capture causal interdependencies between factors and predict how the effects of different inputs propagate through the pathways of the bioprocess mechanisms. This model is interpretable and enables the design of process control policies that are robust against model risk. We present a computationally efficient, provably convergent stochastic gradient method for optimizing such policies. Validation is conducted on a realistic application with a multidimensional, continuous state variable. History: Accepted by Bruno Tuffin, Area Editor for Simulation. Funding: This work was partially supported by National Institute of Standards and Technology [Grant 70NANB17H002]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijoc.2022.1232 .

We consider the problem of deriving confidence intervals for the mean response of a system that is represented by a stochastic simulation whose parametric input models have been estimated from “real-world” data. As opposed to standard simulation confidence intervals, we provide confidence intervals that account for uncertainty about the input model parameters; our method is appropriate when enough simulation effort can be expended to make simulation-estimation error relatively small. To achieve this we introduce metamodel-assisted bootstrapping that propagates input variability through to the simulation response via an equation-based model rather than by simulating. We develop a metamodel strategy and associated experiment design method that avoid the need for low-order approximation to the response and that minimizes the impact of intrinsic (simulation) error on confidence level accuracy. Asymptotic analysis and empirical tests over a wide range of simulation effort show that confidence intervals obtained via metamodel-assisted bootstrapping achieve the desired coverage.