This study examines the scheduling challenges faced by a real-world biomanufacturing site. This fermentation factory is a multi-product, multi-stage facility where batches are scheduled to meet customer demand. Importantly, in the production process of these batches, they are split and mixed into subbatches. Between stages, the intermediate product’s expiration should be avoided by maintaining a continuous flow without waiting in the factory, resulting in zero-wait policies. Sequence-dependent cleaning operations are also required between processing steps to avoid cross-contamination. Recent advances in constraint programming (CP) have demonstrated strong performance on standard scheduling problems. In this work, we investigate how effectively the biomanufacturing scheduling problems can be modeled with CP. Then, we investigate the performance of state-of-the-art CP solvers for solving real, large-scale biomanufacturing scheduling instances. Our empirical evaluation shows that CP Optimizer is more effective than Google OR Tools CP SAT. We show that a warm-start strategy based on domain knowledge improves the performance of the CP approach. Our empirical results show that relaxing the zero-wait constraints results in lower optimality gaps. Finally, we make our set of problem instances and CP model publicly available, thereby extending the scheduling literature with large-scale, realistic benchmarking instances obtained by our industrial collaboration.

When optimizing problems with uncertain parameter values in a linear objective, decision-focused learning enables end-to-end learning of these values. We are interested in a stochastic scheduling problem, in which processing times are uncertain, which brings uncertain values in the constraints, and thus repair of an initial schedule may be needed. Historical realizations of the stochastic processing times are available. We show how existing decision-focused learning techniques based on stochastic smoothing can be adapted to this scheduling problem. We include an extensive experimental evaluation to investigate in which situations decision-focused learning outperforms the state of the art, i.e., scenario-based stochastic optimization.

We study a highly complex scheduling problem that requires the generation and optimization of production schedules for a multi-product biomanufacturing system with continuous and batch processes. There are two main objectives here; makespan and lateness, which are combined into a cost function that is a weighted sum. An additional complexity comes from long horizons considered (up to a full year), yielding problem instances with more than 200 jobs, each consisting of multiple tasks that must be executed in the factory. We investigate whether a rolling-horizon principle is more efficient than a global strategy. We evaluate how cost function weights for makespan and lateness should be set in a rolling-horizon approach where deadlines are used for subproblem definition. We show that the rolling-horizon strategy outperforms a global search, evaluated on problem instances of a real biomanufacturing system, and we show that this result generalizes to problem instances of a synthetic factory.