Machine Learning-enabled Modeling Approach for Predictive Maintenance Decision-making Support

2020 ◽  
Author(s):  
Chunsheng Yang ◽  
Yubin Yang ◽  
Xiaohui Yang ◽  
Qiangqiang Cheng
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Predictive Maintenance ◽  
Modeling Approach ◽  
Maintenance Decision ◽  
Decision Making Support

scholarly journals Real-Time AI-Based Informational Decision-Making Support System Utilizing Dynamic Text Sources

2021 ◽  
Vol 11 (13) ◽  
pp. 6237
Author(s):  
Azharul Islam ◽  
KyungHi Chang
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Random Forest ◽  
Support System ◽  
Classification Accuracy ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Unstructured Data ◽  
Stochastic Gradient Descent ◽  
Decision Making Support

Unstructured data from the internet constitute large sources of information, which need to be formatted in a user-friendly way. This research develops a model that classifies unstructured data from data mining into labeled data, and builds an informational and decision-making support system (DMSS). We often have assortments of information collected by mining data from various sources, where the key challenge is to extract valuable information. We observe substantial classification accuracy enhancement for our datasets with both machine learning and deep learning algorithms. The highest classification accuracy (99% in training, 96% in testing) was achieved from a Covid corpus which is processed by using a long short-term memory (LSTM). Furthermore, we conducted tests on large datasets relevant to the Disaster corpus, with an LSTM classification accuracy of 98%. In addition, random forest (RF), a machine learning algorithm, provides a reasonable 84% accuracy. This research’s main objective is to increase the application’s robustness by integrating intelligence into the developed DMSS, which provides insight into the user’s intent, despite dealing with a noisy dataset. Our designed model selects the random forest and stochastic gradient descent (SGD) algorithms’ F1 score, where the RF method outperforms by improving accuracy by 2% (to 83% from 81%) compared with a conventional method.

An integrated machine learning: Utility theory framework for real-time predictive maintenance in pumping systems

2020 ◽  
pp. 095440542097051
Author(s):  
Raghad M Khorsheed ◽  
Omer Faruk Beyca
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Utility Theory ◽  
Predictive Maintenance ◽  
Manufacturing Operations ◽  
Pumping Stations ◽  
Bearing Failures ◽  
Temperature Levels ◽  
The Right

Bearings are the most widely used mechanical parts in rotating machinery under high load and high rotational speeds. Operating continuously under such harsh conditions, wear and failure are imminent. Developing defects give rise to even-higher vibration and temperature levels. In general, mechanical defects in a machine cause high vibration levels. Therefore, bearing fault identification and early detection enables the maintenance team to repair the problem before it triggers catastrophic failure in the bearing. Machine downtime is thus avoided or minimized. This paper explores the use of Machine Learning (ML) integrated with decision-making techniques to predict possible bearing failures and improve the overall manufacturing operations by applying the correct maintenance actions at the right time. The accuracy of the Predictive Maintenance (PdM) module has been tested on real industrial production datasets. The paper proposes an effective PdM methodology using different ML algorithms to detect failures before they happen and reduce pump downtime. The performance of the tested ML algorithms is based on five performance indicators: accuracy, precision, F-score, recall, and an area under curve (AUC). Experimental results revealed that all tested ML algorithms are successful and effective. Furthermore, decision making with utility theory has been employed to exploit the probability of failures and thus help to perform the appropriate maintenance interventions. This provides a logical framework for decision-makers to identify the optimum action with the maximum expected benefit. As a case study, the model is applied on forwarding pumping stations belonging to the Sewerage Treatment Company (STC), one of the largest sewage stations in Qatar.

Key Techniques in Intelligent Predictive Maintenance (IPdM) – A Framework of Intelligent Faults Diagnosis and Prognosis System (IFDaPS)

2014 ◽  
Vol 1039 ◽  
pp. 490-505 ◽  
Author(s):  
Ke Sheng Wang
Keyword(s):  
Decision Making ◽  
Frequency Domain ◽  
Wavelet Packet ◽  
Remaining Useful Life ◽  
Predictive Maintenance ◽  
Ann Model ◽  
Diagnosis And Prognosis ◽  
Faults Diagnosis ◽  
Maintenance Decision

Intelligent predictive maintenance (IPdM) is a maintenance strategy that makes maintenance decisions automatically and dynamically based on Artificial Intelligence and Data mining techniques through condition monitoring of machines, equipment and production processes. IPdM system consists of the following main modules: sensor and data acquisition, signal and data processing, feature extractions, maintenance decision-making, key performance indicators, maintenance scheduling optimization and feedback control and compensation. Among them, the most important part of IPdM is maintenance decision-making, which includes diagnostics and prognostics. This paper proposes a framework of intelligent faults diagnosis and prognosis system (IFDaPS) and discuss some key technologies for implement IPdM policy in manufacturing and industries. A case study focus on the vibration signals collected from the sensors mounted on a pressure blower for critical components monitoring. We decompose the pre-processed signals into several signals using Wavelet Packet Decomposition (WPD), and then the signals are transformed to frequency domain using Fast Fourier Transform (FFT). The features extracted from frequency domain are used to train Artificial Neural Network (ANN). Trained ANN model is able to identify the fault of the components and predict its Remaining Useful Life (RUL). The case study demonstrates how to implement the proposed framework and intelligent technologies for IPdM and the result indicates its higher efficiency and effectiveness comparing to traditional methods.

scholarly journals Intelligent decision-making support system for manufacturing solution recommendation in a cloud framework

2020 ◽  
Author(s):  
Alessandro Simeone ◽  
Yunfeng Zeng ◽  
Alessandra Caggiano
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Support System ◽  
Cloud Manufacturing ◽  
Cloud Platform ◽  
Intelligent Decision Making ◽  
System A ◽  
Intelligent Decision ◽  
Learning Procedure ◽  
Decision Making Support

AbstractCloud manufacturing represents a valuable tool to enable wide sharing of manufacturing services and solutions by connecting suppliers and customers in large-scale manufacturing networks through a cloud platform. In this context, with increasing manufacturing network size at global scale, the elevated number of manufacturing solutions offered via cloud platform to connected customers can increase the complexity of decision-making, resulting in poor user experience from a customer perspective. To tackle this issue, in this paper, an intelligent decision-making support tool based on a manufacturing service recommendation system (RS) is designed and developed to provide for tailored manufacturing solution recommendation to customers in a cloud manufacturing system. A machine learning procedure based on neural networks for data regression is employed to process historical data on user manufacturing solution preferences and to carry out the automatic extraction of key features from incoming user instances and compatible manufacturing solutions generated by the cloud platform. In this way, the machine learning procedure is able to perform a customer segmentation and build a recommendation list characterized by a ranking of manufacturing solutions which is tailored to the specific customer profile. With the aim to validate the proposed intelligent decision-making support system, a case study is simulated within the framework of a cloud manufacturing platform delivering dynamic sharing of sheet metal cutting manufacturing solutions. The system capability is discussed in terms of machine learning performance as well as industrial applicability and user selection likelihood.

A Risk-Based Maintenance Decision-Making Approach for Railway Asset Management

2018 ◽  
Vol 28 (04) ◽  
pp. 453-483 ◽  
Author(s):  
Li Wang ◽  
Min An ◽  
Yong Qin ◽  
Limin Jia
Keyword(s):  
Decision Making ◽  
Asset Management ◽  
Risk Model ◽  
Cost Model ◽  
Fuzzy Ahp ◽  
Maintenance Cost ◽  
Modeling Methodology ◽  
Maintenance Decision ◽  
Support Model ◽  
Decision Making Support

This paper presents a risk-based maintenance decision making modeling methodology for railway asset maintenance optimization, which takes risk and maintenance cost objectives into consideration in the decision making process. A bottom-up risk analysis approach has been developed by using fuzzy reasoning approach (FRA) and fuzzy-analytical hierarchy process (Fuzzy-AHP) to produce a risk model. A total cost model has also been developed to estimate repair/renewal, maintenance and performance review costs. A risk-based maintenance decision making support model has then been developed by integrating the risk model with cost model in which multi-criteria decision making (MCDM) techniques are employed to process the proposed risk-based maintenance decision making support model. An illustrative example on a section of a track system maintenance decision selection is used to demonstrate the application of the proposed methodology. The results show that by using the proposed methodology the qualitative and quantitative risk data and information with maintenance costs associated with railway assets can be evaluated efficiently and effectively, which provide very useful information to railway engineers, managers, and decision makers.

A risk-based modelling approach to maintenance optimization of railway rolling stock

2019 ◽  
Vol 25 (2) ◽  
pp. 272-293
Author(s):  
Fateme Dinmohammadi
Keyword(s):  
Decision Making ◽  
Cost Function ◽  
Rolling Stock ◽  
Maintenance Optimization ◽  
Content Type ◽  
Modeling Approach ◽  
Inspection And Maintenance ◽  
Maintenance Decision ◽  
Tools And Techniques ◽  
Railway Rolling Stock

Purpose Railway transport maintenance plays an important role in delivering safe, reliable and competitive transport services. An appropriate maintenance strategy not only reduces the assets’ lifecycle cost, but also will ensure high standards of safety and comfort for rail passengers and workers. In recent years, the majority of studies have been focused on the application of risk-based tools and techniques to maintenance decision making of railway infrastructure assets (such as tracks, bridges, etc.). The purpose of this paper is to present a risk-based modeling approach for the inspection and maintenance optimization of railway rolling stock components. Design/methodology/approach All the “potential failure modes and root causes” related to rolling stock systems are identified from an extensive literature review followed by an expert’s panel assessment. The failure causes are categorized into six groups of electrical faults, structural damages, functional failures, degradation, human errors and natural (external) hazards. Stochastic models are then proposed to estimate the likelihood (probability) of occurrence of a failure in the rolling stock system. The consequences of failures are also modeled by an “inflated cost function” that involves safety-related costs, corrective maintenance and renewal (M&R) costs, the penalty charges due to train delays or service interruptions as well as the costs associated with loss of reputation (or loss of fares) in the case of trip cancellation. Lastly, a time-varying risk-cost function is formulated to determine the optimal frequency of preventive inspection and maintenance actions for rolling stock components. Findings For the purpose of clearly illustrating the proposed risk-based inspection and maintenance modeling methodology, a case study of the Class 380 train’s pantograph system from a Scottish train operating company is provided. The results indicate that the proposed model has a substantial potential to reduce the M&R costs while ensuring a higher level of safety and service quality compared to the currently used inspection methodologies. Practical implications The railway rolling stocks should be regularly inspected and maintained so as to ensure network availability and reliability, passenger safety and comfort, and operations efficiency. Despite the best efforts of the maintenance staff, it is reported that a considerable amount of maintenance resources (e.g. budget, time, manpower) is wasted due to insufficiency or inefficiency of current periodic M&R interventions. The model presented in this paper helps the maintenance engineers to assess the current maintenance practices and propose or initiate improvement actions when needed. Originality/value There are few studies investigating the application of risk-based tools and techniques to inspection and maintenance decision making of railway rolling stock components. This paper presents a modeling approach aimed at planning the preventive repair and maintenance interventions for rolling stock components based on risk measures. The author’s model is also capable of incorporating real measurement information gathered at each inspection epoch to update future inspection plans.

Sign in / Sign up

Export Citation Format

Share Document