The high level of computerisation of modern production systems has led to the generation of large volumes of machining data, offering significant potential for using data-driven decision making (DDDM) for examining the fault analysis and incorporating efficient preventive maintenance measures. This research proposes a DDDM framework to incorporate data driven preventive maintenance and fault analysis in modern production systems. This framework uses IoT sensors, cloud computing, and machine learning for real-time fault analysis and preventive maintenance. This approach incorporates the analysis of failures using past machining data and suggests maintenance strategies based on fault analysis. The framework exhibits a continuous model improvement nature based on the system data by including continuous monitoring and model refinement loops. The failure mode effect analysis does systematic fault identification, predictive maintenance planning, and continuous improvement through feedback loops. The applicability of the framework is demonstrated through an experimental study on CNC turning of D2 alloy steel performed on a FANUC CNC lathe. The systematic fault analysis, visualisation of tool wear with machining parameters, capacity utilisation analysis, tool life prediction by multivariate linear regression machine learning model, failure analysis and FMEA were performed by the deep analysis of the recorded experimental data. A random forest model demonstrated an accuracy of 84.37% in tool failure prediction, effectively correlating predicted and actual tool life. Thus, by proposing framework and demonstrating its applicability in an industrial setup, this research establish theoretical as well as practical foundation for applying data driven preventive maintenance and fault analysis in modern production systems to enhance production quality and efficiency.

The paper aims to analyze and optimize the FSWing process of plates in a butt configuration. The ultimate objective is to desensitize the process to both endogenous and exogenous noises, which are lacing the manufacturing process.. To this end, the Robust Design for Products and Processes using the stochastic frontier (RDPP-SF) and the Taguchi-Grey methods are employed, and settings of the process parameters, which yield the best mechanical performances, are sought. The process parameters are the pin rotation speed (x 1 ), feed rate (x 2 ), and plunged surface (x 3 ), whereas the process responses are the longitudinal stress (σ L ), transverse stress (σ T ), longitudinal strain (ε L ), and transverse strain (ε T ) that occur in the FSWed seams. Three replications of an FCCD design are conducted and then statistically analyzed. For the RDPP-SF method, the optimal mechanical properties are σ L = 22.197 MPa, σ T = 22.607 MPa, ε L = 18.33 %, and ε T = 15.83 % and they occur when setting the rotation speed (x 1 ) at 1700 rpm, the feed rate (x 2 ) at 29 mm/min, and the plunged surface (x 3 ) at 478 mm 2 . The optimal values with regard to the Taguchi-Grey method are achieved at, σ L = 23.733 MPa, σ T = 23.886 MPa, ε L = 17.76 %, and ε T = 16.91 % while setting the rotation speed (x 1 ) at 1280 rpm, the feed rate (x 2 ) at 29 mm/min, and the plunged surface (x 3 ) at 478 mm 2 . The discrepancy between the RDPP-SF and the Taguchi-Grey method is partly due to the variance decomposition models, which are employed.

Chatter remains a critical limitation in milling operations, impairing surface integrity, reducing tool life, and limiting productivity. Accurate real-time prediction of chatter is essential for achieving process stability and enhancing machining efficiency. This paper provides a comparative evaluation of Support Vector Machines (SVM) and Artificial Neural Networks (ANN) combined with Variational Mode Decomposition (VMD) for the accurate identification of chatter in real-time vibration data. The milling vibration signals are decomposed into an Intrinsic Mode Functions (IMFs) with VMD to obtain chatter sensitive features. Axial depth of cut, table feed and Spindle speed data combined with these characteristics are used as input to both ANN and SVM models. The ANN model uses Tangent Sigmoid (TANSIG) activation function and six training algorithms, out of which Levenberg-Marquardt (LM) is found to produce the best results. Experimental verification proved that the prediction accuracy of 91.44% was outstripped (over 87.21%) in terms of the SVM model by the ANN model. The robustness of this framework in detecting stable, transitional, and unstable cutting zones has also been demonstrated through Stability Lobe Diagrams (SLDs) analysis. The proposed VMD-ANN approach offers an accurate and efficient solution to chatter prediction and parameter optimisation in real-time, which further contributes to higher material removal rates, better surface quality, and a longer tool life for smart manufacturing applications.

Monitoring tool health is essential for maintaining efficiency, productivity, and quality in lathe turning operations. Traditional methods rely on manual assessments and subjective judgments, which can be time-consuming, inconsistent, and inadequate for detecting subtle tool wear. Therefore, this review discusses the literature review on predicting tool wear in the turning process, comprehensively examining the methods documenting for sensing and testing parameter design, image processing, and classification methods. The review outlines the use of vibration signals and images as datasets and advanced artificial intelligence techniques like machine learning, computer vision, deep learning, and expert systems to predict the accurate wear percentage in the tool. It also discusses the benefits and limitations of methods used in reviewed papers. To conclude, the performance of AI techniques from the reviewed papers, RNN from deep learning, gives more accuracy, with 97.04% predicting the tool wear. Within the Industry 4.0 framework, after a detailed review of the AI techniques, the combination of deep learning techniques that ensemble vibration signals and image information develops as a vital technology for evolving intelligent manufacturing.

Quality function deployment (QFD) is a systematic approach to transform customer requirements (CRs) into product engineering characteristics (ECs). Traditional QFD relies on market research or customer questionnaires to collect a series of ambiguous and uncertain CRs. As a result, evaluating the weighting of CRs and determining the design matrix between CRs and ECs have become the focus and difficulty of QFD. This paper proposes the grey system theory in artificial intelligence technology combined with QFD to develop grey-QFD to solve the issues mentioned before. First, collect the average evaluation values between the aesthetic images and customer satisfaction of representative products. The grey prediction GM (1, N) model is used to obtain the weight of aesthetic needs relative to customer satisfaction and import it into the left QFD. Second, the domain experts decomposed the product form into a morphological analysis table, and fuzzy Delphi screened key ECs and imported them into the ceiling of QFD. Finally, grey relationship analysis established the aesthetic product design matrix between CRs and ECs, and calculated and ranked the final weights of each ECs by using grey relationship degree. The research uses the security camera in the smart home as an experimental object. After operating the proposed grey-QFD, the aesthetic quality of the target product (lively, intelligent, friendly, personalized, and fashionable) and the optimization of the corresponding product ECs are obtained. The result provides a theoretical reference for designers and significantly improves customer aesthetic satisfaction.

Concurrent execution of design and construction tasks is an important way to realize the integration of them in design-build (DB) mode, but it may bring about period risk due to multiple rework and frequent information transfer in local scope. To solve this problem, this study constructs a concurrent execution strategy model from the perspective of quantitative analysis with the decision goal of minimizing the total execution duration. The results show that: first, when the probability of information change between design and construction tasks decreases gradually, the optimal overlap degree increases and the optimal parallelism decreases. Second, learning effect can effectively shorten the minimum duration corresponding to the optimal overlap degree under any degree of parallelism. Thirdly, with the decrease of the probability of information change between design and construction tasks and the strengthening of learning effect, the shortening rate of the actual execution duration corresponding to the optimal parallelism will be gradually greater than the increasing rate of the actual execution duration. The findings of the study provide suggestions for exploring the mechanism of concurrent execution in DB mode, and also provide countermeasures for better realization of integration of design and construction in DB mode.

Through the prevalence of sustainable ideas, automobiles are increasingly pursuing environmental protection strategies for green design, the non-traditional hybrid electric vehicles (HEV) are promoted continuously. If the company can add emotional value to the modeling of HEV, it will be helpful to its sustainable design and sales promotion of it. Therefore, an innovative model combining fuzzy linguistic preference relations (FLPR) and fuzzy quality function deployment (QFD) is proposed here to explore the connection between customer sentiment and the front view of the HEV. Compared with the previous methods, FLPR has the advantages of fewer comparison times and high consistency. First, find out the customer’s emotional expectations and attribute weight ranking for HEV through FLPR, and import customer requirements (CRs) on the left side of fuzzy QFD. Second, the grey prediction model was used to screen out the key engineering features (ECs) and the initial weight of HEV. Finally, based on human subjective imprecise natural semantics, fuzzy QFD established a matrix association between CRs and key ECs, then finally obtained the optimal combination of ECs' final weight and morphological design. The results can assist designers to shorten product development cycles and improve customers' emotional satisfaction, which provides a theoretical reference for the sustainable design and marketing of environmentally friendly cars in the future.

In the recent past, Non-Destructive Testing (NDT) has become the most popular technique due to its efficiency and accuracy without destroying the object and maintaining its original structure and gathering while examining external and internal welding defects. Generally, the NDT environment is harmful which is distinguished by huge volatile fields of electromagnetic, elevated radiation emission instability, and elevated heat. Therefore, a suitable NDT approach could be recognized and practiced. In this paper, a novel algorithm is proposed based on a Phased array ultrasonic test (PAUT) for NDT to attain the proper test attributes. In the proposed methodology, the carbon steel welding section is synthetically produced with various defects and tested using the PAUT method. The signals which are acquired from the PAUT device are having noise. The Adaptive Least Mean Square (ALMS) filter is proposed to filter PAUT signal to eliminate random noise and Gaussian noise. The ALMS filter is the combination of low pass filter (LPF), high pass filter (HPF), and bandpass filter (BPF). The time-domain PAUT signal is converted into a frequency-domain signal to extract more features by applying the Empirical Wavelet Transform (EWT) algorithm. In the frequency domain signal, first order and second order features extraction techniques are applied to extract various features for further classification. The Deep Learning methodology is proposed for the classification of PAUT signals. Based on the PAUT signal features, the Deep Convolution Neural Network (DCNN) is applied for further classification. The DCNN will classify the welding signal as to whether it is defective or non-defective. The Confusion Matrix (CM) is used for the estimation of measurement of performance of classification as calculating accuracy, sensitivity, and specificity. The experiments prove that the proposed methodology for PAUT testing for welding defect classification is obtained more accurately and efficiently across existing methodologies by providing numerical and graphical results.