Scope for Machine Learning in Digital Manufacturing

This provocation paper provides an overview of the underlying optimisation problem in the emerging field of Digital Manufacturing. Initially, this paper discusses how the notion of “Digital Manufacturing” is transforming from a term describing a suite of software tools for the integration of production and design functions towards a more general concept incorporating computerised manufacturing and supply chain processes, as well as information collection and utilisation across the product life cycle. On this basis, we use the example of one such manufacturing process, Additive Manufacturing, to identify an integrated multi-objective optimisation problem underlying Digital Manufacturing. Forming an opportunity for a concurrent application of data science and optimisation, a set of challenges arising from this problem is outlined. The emergence of Digital Manufacturing In manufacturing, the concept of Digital Manufacturing has arisen and evolved over the recent decades. Initially known as Computer-Integrated Manufacturing, the concept traditionally describes the utilisation of a suite of tools to facilitate the integration of product and process design, with a particular emphasis on jointly optimising “manufacturing before starting the production and supporting ramp-up phases” (Chryssolouris et al., 2009). Cutting across the engineering and operations functions, this collection of digital tools supports process and tooling design, plant layout, advanced visualisation, simulation and concurrent engineering approaches (Slansky, 2008). Broadly, the traditional understanding of Digital Manufacturing can be viewed as part of a change from cost-driven to knowledge-based manufacturing (Westkämper et al., 2007). Following the emergence and strongly growing relevance of the concept of “Industrie 4.0”, the flavour of Digital Manufacturing is changing, however. With a much greater focus on the utilisation of real time data in closed loop control architectures obtained via ubiquitous sensing and computing, the optimisation of 1 3D Printing Research Group, Faculty of Engineering, email: [email protected] 2 Automated Scheduling and Planning Research Group, School of Computer Science, email: [email protected] networked production facilities has now taken centre stage in Digital Manufacturing. Similarly, the evolved understanding of Digital Manufacturing places a much greater emphasis on flexibility, reconfigurability and resilience in the operation of manufacturing systems (Siemens, 2014). Also described as the application of cyber-physical systems in manufacturing, such systems are based on the idea of creating digital models of processes and products that are expanded throughout various stages in the production flow and later stages in the product life cycle. Innovation in underlying manufacturing processes However, the emergence of Digital Manufacturing has also resulted in innovation within the underlying manufacturing processes themselves, which remain the centre of manufacturing innovation (Westkämper, 2007). Among other aspects, such process innovation promises to remove existing technological tradeoffs in manufacturing, permit novel supply chain configurations and enable new value propositions in manufacturing.