Data analytics pipeline for prediction and decision making in complex products and systems development

Abstract

Facilitating data analytics for effective prediction in complex products or systems development is the focus of the research described in this paper. The specific objective was to develop strategies and a data analytics pipeline with a view to supporting exploration of the design space of complex products or systems upfront. The underlying challenges tackled included how to acquire and store raw data gathered by using both the traditional methods and advanced Internet of Things (IoT) devices, how to preprocess and transform raw data into a form suited for data analytics, and how to deal with analytics. A pipeline for data analytics to support decision making in complex products or systems development is proposed and its applicability illustrated with a practical example. The incorporation of advanced analytics techniques into the proposed pipeline allows users to acquire data and to insightfully and intelligently predict aspects such as cost and assembly time early on, and to make decisions based on data that may otherwise deemed to be inaccessible or unusable. This work contributes to the efforts directed toward applying data analytics techniques in a way that can have a profound impact on an engineering product or system development process.