An analytical framework for smart manufacturing

Abstract

Smart manufacturing is an emerging paradigm for the next generation of manufacturing systems. One key to the success of smart manufacturing is the ability to use the production data for defining predictive and descriptive models and their analyses. However, the development and refinement of such models is a labor- and knowledge-intensive activity that involves acquiring data, selecting and refining an analytical method and validating results. This paper presents an analytical framework that facilitates these activities by allowing ad-hoc analyses to be rapidly specified and performed. The proposed framework uses a domain-specific language to allow manufacturing experts to specify analysis models in familiar terms and includes code generators that automatically generate the lower-level artifacts needed for performing the analysis. We also describe the use of our framework with an example problem.