Applying Predictive Maintenance in Flexible Manufacturing

Abstract

In Industry 4.0 context, manufacturing related processes e.g. design processes, maintenance processes are collaboratively processed across different factories and enterprises. The state i.e. operation, failures of production equipment tools could easily impact on the collaboration and related processes. This complex collaboration requires a flexible and extensible system architecture and platform, to support dynamic collaborations with advanced capabilities such as big data analytics for maintenance. As such, this paper looks at how to support data-driven and flexible predictive maintenance in collaboration using FIWARE? Especially, applying big data analytics and datadriven approach for effective maintenance schedule plan, employing FIWARE Framework, which leads to support collaboration among different organizations modularizing of different related functions and security requirements.

A Predictive Maintenance Model for Flexible Manufacturing in the Context of Industry 4.0

“…One of the challenges for Industry 4.0 manufacturing is to design and develop embedded services assisting in a flexible way the effective management of machine equipment tools by reducing downtimes and costs (Mobley 2002;Russell and Norvig 2010;Lee and Kao, 2014;Sang et al, 2020;Zonta et al, 2020;Sang et al, 2021). Our work focused on the design and development of a predictive maintenance model for Industry 4.0 (i.e., Predictive Maintenance for Industry 4.0) which utilizes the proposed predictive maintenance scheduling for multiple machine components (i.e., Predictive Maintenance Schedule for Multiple Machines and Components) by taking into account machine data such as operation, condition, and maintenance data.…”

“…For traditional manufacturing organization, resource dependency may not be as critical as the Industry 4.0 collaborative aspect, since the traditional organization does not need coordination or data outside its own organization, and it can probably has its own capable resource. For an effective predictive maintenance, these resource dependencies must be considered, especially for scheduling (Sang et al, 2021). For developing the predictive RUL models, a similar type of machine equipment tool is required (Zheng et al, 2017).…”

“…Handling the demands, i.e., highly collaborative complex systems and multiple machines of Industry 4.0, focusing on manufacturing is challenging (Thoben et al, 2017). From the requirements of Industry 4.0 in the literature studies (Lee et al, 2015a;Thoben et al, 2017;Sang et al, 2020aSang et al, , 2020bSang et al, 2020;Zonta et al, 2020;Sang et al, 2021), several key factors need to be considered for an optimal maintenance schedule plan of PMS4MMC. This includes data-driven maintenance, i.e., predictive models such as RUL, multiple machine components, maintenance tasks, maintenance time, cost, and the resource aspect, i.e., availability status of each component and engineer.…”

“…Predictive maintenance aids in the effective management of industrial assets. It provides a diagnosis of faults in related different machinery, deficient processes, and detailed inspection of the detection using various analyses (Mobley 2002;Sang et al, 2020b). Predictive maintenance involves two key components: inquisition of knowledge through analytics, i.e., prediction and detection of machine tools, and decision-supported schedule plan for the required maintenance task to be completed.…”

Supporting Predictive Maintenance in Virtual Factory

“…integrated processes, value and cost associated with downtime, faulty products, etc. [7,8]. Data generated by the various processes, systems/machine equipment tools across factories operation and production offer opportunities such as data-driven analytics e.g.…”