A novel predict-prevention quality control method of multi-stage manufacturing process towards zero defect manufacturing

Zhao, Liping; Li, Bo-Hao; Yao, Yiyong

doi:10.1007/s40436-022-00427-9

Cited by 9 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The influence of each factor on the final ZDM implementation cost, as revealed by our ANOVA analysis, brings forward a crucial aspect for consideration. These findings support the conclusions of [31] regarding the importance of understanding both the individual and the interactive effects of various factors in ZDM implementation. By identifying the most influential factors, manufacturers can channel their resources more efficiently, highlighting areas where they should pay greater attention.…”

Section: Results Discussionsupporting

confidence: 86%

“…Previous studies have underscored the importance of understanding the distinguishing aspects of these two ZDM approaches, but they stop short of providing a detailed model to guide their selection and application in diverse manufacturing settings [30]. The limitations in existing studies manifest in the form of a gap in comprehensive understanding and the lack of a decision-making tool for selecting the most suitable approach [31]. This research work aims to fill this gap by providing an analysis and comparison of these two ZDM approaches and developing a cost model to assist decision-making.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Analysis for Mapping Product-Oriented and Process-Oriented Zero-Defect Manufacturing (ZDM) in the Industry 4.0 Era

Psarommatis

May

2023

Sustainability

View full text Add to dashboard Cite

Quality is a key aspect in the era of Industry 4.0. Zero-defect manufacturing (ZDM) as the latest quality assurance approach. It can be implemented in two different approaches: the product-oriented and the process-oriented ZDM. It is important to know how and when to consider adopting one approach over the other. To achieve that there is the need for analyzing the differences of the two ZDM approaches. However, the current literature lacks a detailed analysis and comparison of these two approaches to ZDM implementation. Earlier studies on the topic have adopted one of these approaches over the other without evaluating how it fits with specific cases. The literature of the last decade indicates a movement towards product-oriented approaches, but it has not shown proof why product oriented was used over process oriented. Guided by these gaps, this research work creates a model for quantifying the effects of the implementation of both the product-oriented and process-oriented ZDM approaches. The proposed model considers all the critical parameters that affect the problem and serves as an assisting tool to engineers during the design or re-configure manufacturing systems, for choosing the most efficient ZDM approach for their specific cases. The robustness of the model was analyzed using the design of experiments method. The results from both the designed experiments and an industrial use case illustrate that in most cases, product-oriented ZDM performs better than the process-oriented approach. Nevertheless, in our analysis, we also highlight strong interactions between some factors that make the selection between product-oriented and process-oriented ZDM difficult and complex.

show abstract

Section: Results Discussionsupporting

confidence: 86%

Section: Introductionmentioning

confidence: 99%

A Systematic Analysis for Mapping Product-Oriented and Process-Oriented Zero-Defect Manufacturing (ZDM) in the Industry 4.0 Era

Psarommatis

May

2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…This provided more efficient data mining, which is crucial to implementing the ZDM concept based on artificial intelligence techniques, such as machine learning. In fact, implementing ZDM can be separated into three main parts: characteristics monitoring, key quality prediction, and assembly quality optimisation [30], fields where AI and specially ML have made significant progress.…”

Section: Zero Defect Manufacturingmentioning

confidence: 99%

Application of Artificial Intelligence Techniques to a Sensorial Module

Pimentel,

Gonçalves,

Pinto

et al. 2024

2024 10th International Conference on Mechatronics and Robotics Engineering (ICMRE)

View full text Add to dashboard Cite

Nowadays, society is in a climate of uncertainty regarding energy and resource sustainability, which is why efforts to combat this problem are made and encouraged daily. This is no different in the industry, where there is a continuous work to improve manufacturing processes in order to achieve utopian zero defect manufacturing. Alongside this, the emergence of new technologies, the development of the computing power of machines and the digitisation of processes in the context of Industry 4.0 have been fundamental to the production of data that feeds and develops artificial intelligence models.This dissertation will address a quality control problem in a welding process in the automotive industry. It is believed that the development of a predictive quality control system using artificial intelligence techniques can solve the problem in question.In order to study the problem, and due to the impossibility of working with the real problem, this work includes a study of the techniques expected to be applied in the final solution. As well as being theoretical, this study of methods has a practical component, applying them to data collected from a testbed. This testbed is a physical prototype with a parameterised process, also developed as part of this dissertation. In addition, a dataset referring to a laser welding process was also analysed to complement the study of the algorithms with data similar to data from the original problem.In this document, the scope of the work is framed within the current state of the art, followed by a theoretical reflection on the subjects covered. It is decided to study and apply the following supervised learning techniques to the data collected: K-Nearest Neighbours, Support Vector Machine, Linear Regression, Ridge Regression, Lasso Regression, Random Forest and Gradient Boosting. This is followed by a description of the test base developed and the methodology for collecting and processing the data.It is concluded that although the models built with the Support Vector Machine algorithms showed satisfactory results for modelling the laser welding process, the data collected from the test base was unreliable, and different models can perform differently in different situations.

show abstract

“…This is also one of the main goals of the proposed solution named Orion (analyzed in Section 3). In this work [13], a predictive quality control method in a multi-stage manufacturing process (MMP) aiming at ZDM is presented. By considering the correlations between different measurements, a deep supervised network is built to predict quality characteristics.…”

Section: Related Researchmentioning

confidence: 99%

Adaptive Quality Diagnosis Framework for Production Lines in a Smart Manufacturing Environment

et al. 2023

View full text Add to dashboard Cite

Production lines in manufacturing environments benefit from quality diagnosis methods based on learning techniques since their ability to adapt to the runtime conditions improves performance, and at the same time, difficult computational problems can be solved in real time. Predicting the divergence of a product’s physical parameters from an acceptable range of values in a manufacturing line is a process that can assist in delivering consistent and high-quality output. Costs are saved by avoiding bursts of defective products in the pipeline’s output. An innovative framework for the early detection of a product’s physical parameter divergence from a specified quality range is designed and evaluated in this study. This framework is based on learning automata to find the sequences of variables that have the highest impact on the automated sensor measurements that describe the environmental conditions in the production line. It is shown by elaborate evaluation that complexity is reduced and results close to optimal are feasible, rendering the framework suitable for deployment in practice.

show abstract

A novel predict-prevention quality control method of multi-stage manufacturing process towards zero defect manufacturing

Cited by 9 publications

References 31 publications

A Systematic Analysis for Mapping Product-Oriented and Process-Oriented Zero-Defect Manufacturing (ZDM) in the Industry 4.0 Era

A Systematic Analysis for Mapping Product-Oriented and Process-Oriented Zero-Defect Manufacturing (ZDM) in the Industry 4.0 Era

Application of Artificial Intelligence Techniques to a Sensorial Module

Adaptive Quality Diagnosis Framework for Production Lines in a Smart Manufacturing Environment

Contact Info

Product

Resources

About