Workspace for industry 4.0

Abstract

Compared to writing software from scratch, Scientific Workflow Systems (SWSs) provide a more convenient way to acquire data, customize inputs and combine algorithms in order to develop reproducible outputs. Workspace is an SWS that has been under continuous development at CSIRO since 2005 and which attempts to address key challenges faced by the research and commercial communities; researcher/developer productivity, reproducibility of results, collaboration between research teams and portability and interoperability between different computing environments and scientific domains. One such domain is that of Industry 4.0, commonly referred to as the "fourth industrial revolution". Industry 4.0 centers on the development of "smart factories" where cyber-physical systems and physical processes operate in concert to create a virtual copy (or "digital twin") of the physical world in order to enable (or in many cases automate) decentralized decision making. In such smart factories, manufacturing processes often take the form of a workflow, where devices, data and algorithms are combined in different ways to produce outputs for decision making and automation purposes. In this paper, we explore how Workspace, in both design and implementation phases, can provide significant value when creating digital twins using arrays of sensors. An Industry 4.0 demonstrator application is used to highlight challenges and to show how a workflow approach can be used in applications representative of real-world needs. The demonstrator application is specifically focused around the inspection of manufactured components for quality-assurance purposes. In doing so, we will analyse how Workspace addresses the specific challenges that are presented by problems within the Industry 4.0 domain. These include the need to tailor the system to specific manufacturing processes at low cost, the ability to cope with large volumes of streaming data and changes in how the data is integrated and interpreted, developing the skills of operators to cope with the complicated processes involved while maintaining existing know-how, and the need to maintain the integrity of the disparate processes. We provide a detailed study of those features and plugins of Workspace which are beneficial to the development of Industry 4.0 applications, such as distributed execution, OpenCV, PCL, ZeroMQ plugins, and the ability to create custom applications. These capabilities, along with the Workspace design principles of analyse, collaborate, commercialise, everywhere, align it well with problems in the Industry 4.0 space, allowing the creation of high performing implementations for use in industrial settings.