[Paper Review] State-of-the-Art Review: The Use of Digital Twins to Support Artificial Intelligence-Guided Predictive Maintenance

Sizhe Ma, Katherine A. Flanigan+1 more|arXiv (Cornell University)|Jun 19, 2024

Machine Fault Diagnosis Techniques6 citations

TL;DR

This paper provides a roadmap and literature review on integrating Digital Twins with AI-guided Predictive Maintenance (PMx), identifying information/functional requirements, existing DT implementations, gaps, and future research directions.

ABSTRACT

In recent years, predictive maintenance (PMx) has gained prominence for its potential to enhance efficiency, automation, accuracy, and cost-effectiveness while reducing human involvement. Importantly, PMx has evolved in tandem with digital advancements, such as Big Data and the Internet of Things (IOT). These technological strides have enabled Artificial Intelligence (AI) to revolutionize PMx processes, with increasing capacities for real-time automation of monitoring, analysis, and prediction tasks. However, PMx still faces challenges such as poor explainability and sample inefficiency in data-driven methods and high complexity in physics-based models, hindering broader adoption. This paper posits that Digital Twins (DTs) can be integrated into PMx to overcome these challenges, paving the way for more automated PMx applications across various stakeholders. Despite their potential, current DTs have not fully matured to bridge existing gaps. Our paper provides a comprehensive roadmap for DT evolution, addressing current limitations to foster large-scale automated PMx progression. We structure our approach in three stages: First, we reference prior work where we identified and defined the Information Requirements (IRs) and Functional Requirements (FRs) for PMx, forming the blueprint for a unified framework. Second, we conduct a literature review to assess current DT applications integrating these IRs and FRs, revealing standardized DT models and tools that support automated PMx. Lastly, we highlight gaps in current DT implementations, particularly those IRs and FRs not fully supported, and outline the necessary components for a comprehensive, automated PMx system. Our paper concludes with research directions aimed at seamlessly integrating DTs into the PMx paradigm to achieve this ambitious vision.

Motivation & Objective

• Identify and formalize the information requirements (IRs) and functional requirements (FRs) for predictive maintenance (PMx).
• Survey how Digital Twins (DTs) are currently used to satisfy these IRs/FRs.
• Identify standardized DT models and tools that support automated PMx.
• Highlight gaps where current DTs do not fully support IRs/FRs and delineate components for a comprehensive PMx system.
• Propose a research agenda for integrating DTs into PMx to enable large-scale automated PMx.

Proposed method

• Build on prior work to define IRs and FRs for PMx as a foundational blueprint.
• Conduct an exhaustive literature review across disciplines to assess current DT usage in PMx.
• Identify standardized DT models, tools, and environments that support automated PMx.
• Analyze gaps where IRs/FRs are not fully supported by DTs.
• Delineate components and a roadmap for a comprehensive, automated PMx system using DTs.

Experimental results

Research questions

• RQ1 What is the current status of Digital Twins in supporting AI-guided PMx?
• RQ2 Which IRs and FRs identified for PMx are currently fulfilled by DT implementations?
• RQ3 What standardized DT models and environments exist that can facilitate automated PMx?
• RQ4 What gaps remain in DT capabilities to fully realize automated PMx, and what components are needed to bridge them?
• RQ5 What future research directions are needed to integrate DTs into PMx at scale?

Key findings

• DTs have transformative potential for PMx but current state-of-the-art DTs are not yet mature enough for widespread industrial adoption.
• DT frameworks comprise multiple components (PT, PTE, DT, DTE, DE, instrumentation, realization, DThread, HR, analysis, accountability, QR) that together enable data flow and decision-making for PMx.
• There are existing standardized models and environments (e.g., point cloud modeling, BIM, FEM, fuzzy logic) that can support various PMx tasks, but their integration into a unified PMx DT is still developing.
• A clear set of IRs and FRs enables systematic evaluation of DTs against PMx needs, but several IRs/FRs remain insufficiently addressed by current DTs.
• The paper provides a roadmap and research directions to advance DTs toward seamless integration with PMx for automated maintenance.
• The analysis emphasizes the need for accountable, interpretable, robust, scalable, and transferable DT capabilities to gain trust and enable broader deployment.