Deep Learning Approaches to Remaining Useful Life Prediction: A Survey

Cummins, Logan; Killen, Brad; Kirby, Thomas J.; Barrett, Paul M.; Rahimi, Shahram; Seale, Maria

doi:10.1109/ssci50451.2021.9659965

Cited by 8 publications

(10 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the past several years, modern ML techniques such as Deep Learning (DL) have garnered much attention in PHM. To this end, [10,11,12,13,14] have reviewed several DL techniques for system health monitoring and management. Likewise, Carvalho et al [15], who reviewed multiple ML algorithms in PdM that use Run to Failure (R2F) data, identified PdM as an emerging tool that helps with scheduling maintenance events.…”

Section: Related Workmentioning

confidence: 99%

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

et al. 2023

View full text Add to dashboard Cite

Early detection of faulty patterns and timely scheduling of maintenance events can minimize risk to the underlying processes and increase a system's lifespan, reliability, and availability. Two main data-driven approaches are used in the literature to determine the Remaining Useful Life (RUL): direct calculation from raw data and indirect analysis by revealing the transitions from one latent state to another and highlighting degradation in a system's Health Indices. The present study discusses the state-of-theart data-driven methods introduced for RUL prediction in predictive maintenance, by looking at their capabilities, scalability, performance, and weaknesses. We will also discuss the challenges faced with the current approaches and the future directions to tackle the current limitations.

show abstract

Section: Related Workmentioning

confidence: 99%

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Over the past several years, modern ML techniques such as Deep Learning (DL) have garnered much attention in PHM. To this end, [10], [11], [12], [13], [14] have reviewed several DL techniques for system health monitoring and management. Likewise, Carvalho et al [15], who reviewed multiple ML algorithms in PdM that use Run to Failure (R2F) data, identified PdM as an emerging tool that helps with scheduling maintenance events.…”

Section: Related Workmentioning

confidence: 99%

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

Ramezani¹,

Cummins²,

Killen³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The initial step of our proposed methodology is to develop a PdM model for RUL estimation. Researchers have already published tons of work in this area, and a brief survey of them can be found in (Zhang, Si, Hu, & Lei, 2018;Lipu et al, 2018;Cummins et al, 2021;Chen, Hong, & Zhou, 2022b).…”

Section: Rul Prediction Model Developmentmentioning

confidence: 99%

“…Data-driven predictive maintenance (PdM) approaches based on black-box machine learning (ML)/deep learning (DL) models have achieved remarkable success in terms of predictive accuracy and capability of modelling complex systems (Cummins et al, 2021;Keleko, Kamsu-Foguem, Ngouna, & Tongne, 2022;Chen, Hong, & Zhou, 2022a;Jayasinghe, Samarasinghe, Yuenv, Low, & Ge, 2019). However, the complete repair plan and maintenance actions that must be performed based on the detected symptoms of damage and wear often require complex reasoning and planning processes involving many actors and balancing different priorities-which cannot be fully automated in many cases.…”

Section: Introductionmentioning

confidence: 99%

Explainable Predictive Maintenance is Not Enough: Quantifying Trust in Remaining Useful Life Estimation

Kundu,

Hoque

2023

PHM_CONF

View full text Add to dashboard Cite

Machine learning (ML)/deep learning (DL) has shown tremendous success in data-driven predictive maintenance (PdM). However, operators and technicians often require insights to understand what is happening, why it is happening, and how to react, which these black-box models cannot provide. This is a major obstacle in adopting PdM as it cannot support experts in making maintenance decisions based on the problems it detects. Motivated by this, several researchers have recently utilized various post-hoc explanation methods and tools, such as LIME, SHAP, etc., for explaining the predicted RUL from these black-box models. Unfortunately, such (post-hoc) explanation methods often suffer from the \emph{disagreement problem}, which occurs when multiple explainable AI (XAI) tools differ in their feature ranking. Hence, explainable PdM models that rely on these methods are not trustworthy, as such unstable explanations may lead to catastrophic consequences in safety-critical PdM applications. This paper proposes a novel framework to address this problem. Specifically, first, we utilize three state-of-the-art explanation methods: LIME, SHAP, and Anchor, to explain the predicted RUL from three ML-based PdM models, namely extreme gradient boosting (XGB), random forest (RF), logistic regression (LR), and one feed-forward neural network (FFNN)-based PdM model using the C-MAPSS dataset. We show that the ranking of dominant features for RUL prediction differs for different explanation methods. Then, we propose a new metric \emph{trust score} for selecting the proper explanation method. This is achieved by evaluating the XAI methods using four evaluation metrics: fidelity, stability, consistency, and identity, and then combining them into a single \emph{trust score} metric through utilizing Kenny and Borda rank aggregation methods. Our results show that the proposed method effectively selects the most appropriate explanation method from a set of explanation methods for estimated RULs. To the best of our knowledge, this is the first work that attempts to address and solve the disagreement problem in explainable PdM.

show abstract

Deep Learning Approaches to Remaining Useful Life Prediction: A Survey

Cited by 8 publications

References 40 publications

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

Explainable Predictive Maintenance is Not Enough: Quantifying Trust in Remaining Useful Life Estimation

Contact Info

Product

Resources

About