Autonomous supervision and optimization of product quality in a multi-stage manufacturing process based on self-adaptive prediction models

•Early recognition of downtrends in product quality by multi-stage ahead prediction models.•Autonomous handling of system dynamics and drifts by flexible adaptive non-linear prediction models.•Non-linear partial least squares regression; methods for incremental, single-pass update of the latent variable space and dynamic forgetting.•Automatic process optimization with the usage of multi-objective evolutionary algorithms and predictive surrogate mappings.•Successful application in a (micro-fluidic) chip production leading to new machine parameter settings for improved product (chip) quality.