With the ongoing digitization of the manufacturing industry and the ability to bring together data from manufacturing processes and quality measurements, there is enormous potential to use machine learning and deep learning techniques for quality assurance. In this context, predictive quality enables manufacturing companies to make data-driven estimations about the product quality based on process data. In the current state of research, numerous approaches to predictive quality exist in a wide variety of use cases and domains. Their applications range from quality predictions during production using sensor data to automated quality inspection in the field based on measurement data. However, there is currently a lack of an overall view of where predictive quality research stands as a whole, what approaches are currently being investigated, and what challenges currently exist. This paper addresses these issues by conducting a comprehensive and systematic review of scientific publications between 2012 and 2021 dealing with predictive quality in manufacturing. The publications are categorized according to the manufacturing processes they address as well as the data bases and machine learning models they use. In this process, key insights into the scope of this field are collected along with gaps and similarities in the solution approaches. Finally, open challenges for predictive quality are derived from the results and an outlook on future research directions to solve them is provided.

Deep learning-based predictive quality enables manufacturing companies to make data-driven predictions of the quality of a produced product based on process data. A central challenge is that production processes are subject to continuous changes such as the manufacturing of new products, with the result that previously trained models may no longer perform well in the process. In this paper, we address this problem and propose a method for continual learning in such predictive quality scenarios. We therefore adapt and extend the memory-aware synapses approach to train an artificial neural network across different product variations. Our evaluation in a real-world regression problem in injection molding shows that the approach successfully prevents the neural network from forgetting of previous tasks and improves the training efficiency for new tasks. Moreover, by extending the approach with the transfer of network weights from similar previous tasks, we significantly improve its data efficiency and performance on sparse data. Our code is publicly available to reproduce our results and build upon them.

The warehousing industry is faced with increasing customer demands and growing global competition. A major factor in the efficient operation of warehouses is the strategic storage location assignment of arriving goods, termed the dynamic storage location assignment problem (DSLAP). This paper presents a real-world use case of the DSLAP, in which deep reinforcement learning (DRL) is used to derive a suitable storage location assignment strategy to decrease transportation costs within the warehouse. The DRL agent is trained on historic data of storage and retrieval operations gathered over one year of operation. The evaluation of the agent on new data of two months shows a 6.3% decrease in incurring costs compared to the currently utilized storage location assignment strategy which is based on manual ABC-classifications. Hence, DRL proves to be a competitive solution alternative for the DSLAP and related problems in the warehousing industry.