Root Cause Of Fault Research Articles

A Comparative Study of Causality Detection Methods in Root Cause Diagnosis: From Industrial Processes to Brain Networks.

Abstracting causal knowledge from process measurements has become an appealing topic for decades, especially for fault root cause analysis (RCA) based on signals recorded by multiple sensors in a complex system. Although many causality detection methods have been developed and applied in different fields, some research communities may have an idiosyncratic implementation of their preferred methods, with limited accessibility to the wider community. Targeting interested experimental researchers and engineers, this paper provides a comprehensive comparison of data-based causality detection methods in root cause diagnosis across two distinct domains. We provide a possible taxonomy of those methods followed by descriptions of the main motivations of those concepts. Of the two cases we investigated, one is a root cause diagnosis of plant-wide oscillations in an industrial process, while the other is the localization of the epileptogenic focus in a human brain network where the connectivity pattern is transient and even more complex. Considering the differences in various causality detection methods, we designed several sets of experiments so that for each case, a total of 11 methods could be appropriately compared under a unified and reasonable evaluation framework. In each case, these methods were implemented separately and in a standard way to infer causal interactions among multiple variables to thus establish the causal network for RCA. From the cross-domain investigation, several findings are presented along with insights into them, including an interpretative pitfall that warrants caution.

Toward understandable semi-supervised learning fault diagnosis of chemical processes based on long short-term memory ladder autoencoder (LSTM-LAE) and self-attention (SA)

Fault diagnosis and localization play vital role in chemical process monitoring. Finding the root cause of fault accurately and timely is the key to avoid serious accidents and ensure process safety. Unfortunately, most deep learning-based models only involves in predicting fault state and interpretability is still not fully explored. Besides, lack of labeled samples in practical situations makes supervised learning difficult to implement. To circumvent the obstacle, semi-supervised learning based on long short-term memory ladder autoencoder is combined with self-attention mechanism, which is intended to establish interpretable model by explicitly clarifying the corresponding relationship between abnormal variables and faults. Using Tennessee Eastman process and practical high-purity carbonate production process as benchmark, fault diagnosis and identification performance of proposed LSTM-LAE-SA is validated, and interpretability analysis is performed to demonstrate its capabilities in abnormal variable locations. The developed understandable model could improve operator's trust and industrial application in fault diagnosis system.

Enhancement of Foley Catheter Assembly Section Quality using Six Sigma (A Case Study of an Indonesian Medical Device Company)

This study aims to enhance the quality of the Foley catheter assembly section at PT XYZ, an Indonesian medical device manufacturer, using Six Sigma methodology. PT XYZ currently faces an 8% defect rate in its Foley catheter assembly section, surpassing the acceptable 3% threshold. The current Sigma level of the Shrink process within the assembly section is 2.36, while the Balloon Vulcanizing and Assembly process stands at 2.68. The Six Sigma DMAIC (Define, Measure, Analyze, Improve, Control) framework was used to identify and address root causes of defects. Data collection methods included interviews, questionnaires, and field observations, revealing key issues such as incorrect tray loading, machine calibration errors, operator skill variability, and inadequate handling techniques. Proposed improvements include updating Standard Operating Procedures (SOPs), implementing SCAMPER for calibration, comprehensive training programs, and incentive schemes. Additional measures involve template overlays for precise cutting, vendor maintenance, and control charts for consistency. The Control phase ensures sustainability through documentation, training, and monitoring using control charts and Failure Modes and Effects Analysis (FMEA). Implementing these solutions is expected to reduce the defect rate and elevate the Sigma level of both processes to 3, enhancing process efficiency and product quality. This will improve operational efficiency and customer satisfaction, providing a robust framework for PT XYZ to elevate its quality control and contribute to better healthcare device manufacturing in Indonesia.

Fault root cause analysis using degree of change and mean variable threshold limit in non-linear dynamic distillation column

The chemical industrial processes are both dynamic and complex. The presence of instability and danger in the production process poses significant challenges to safety management. Early and accurate fault diagnosis is essential to ensure a sustainable and stable chemical process performance. Most studies in literature do not provide detailed information about diagnosing the root cause of faults in distillation columns using contribution plots. Using principal component analysis (PCA) in conventional contribution plots leads to a large number of contributing variables, making it difficult to identify the root cause of a specific fault. In this study, the authors suggest employing the Degree of Change of Contribution Values (DCC) and Mean Variable Threshold Limit (LMVT) techniques alongside the Wavelet Transform-Principal Component Analysis (WT-PCA) method to decrease the quantity of fault identification variables in a non-linear system (distillation column). The Aspen Plus software is utilized to perform a dynamic simulation of a distillation column, generating data for both normal and faulty operations. Two distinct types of faults are considered, which include loss of feed flow and feed temperature changes. This study aims to analyze the results obtained from the Hotelling T2 and SPE contribution plots by calculating the average difference between the normal and faulty conditions using the Degree of Change in Contribution Values. This procedure is performed to determine the extent of deviation from the standard state. Subsequently, the variables that have surpassed the threshold limit for means variables (LMVT) are designated as the variables that serve as the primary underlying cause of the fault. The technique was employed in a case study involving a nonlinear distillation column, and the results indicate that only one "dominant" variable intersects the LMVT by Hotelling T2, which represents the true underlying cause of the corresponding fault.

Chip-Level Defect Analysis with Virtual Bad Wafers Based on Huge Big Data Handling for Semiconductor Production

Semiconductors continue to shrink in die size because of benefits like cost savings, lower power consumption, and improved performance. However, this reduction leads to more defects due to increased inter-cell interference. Among the various defect types, customer-found defects are the most costly. Thus, finding the root cause of customer-found defects has become crucial to the quality of semiconductors. Traditional methods involve analyzing the pathways of many low-yield wafers. Yet, because of the extremely limited number of customer-found defects, obtaining significant results is difficult. After the products are provided to customers, they undergo rigorous testing and selection, leading to a very low defect rate. However, since the timing of defect occurrence varies depending on the environment in which the product is used, the quantity of defective samples is often quite small. Unfortunately, with such a low number of samples, typically 10 or fewer, it becomes impossible to investigate the root cause of wafer-level defects using conventional methods. This paper introduces a novel approach to finding the root cause of these rare defective chips for the first time in the semiconductor industry. Defective wafers are identified using rare customer-found chips and chip-level EDS (Electrical Die Sorting) data, and these newly identified defective wafers are termed vBADs (virtual bad wafers). The performance of root cause analysis is dramatically improved with vBADs. However, the chip-level analysis presented here demands substantial computing power. Therefore, MPP (Massive Parallel Processing) architecture is implemented and optimized to handle large volumes of chip-level data within a large architecture infrastructure that can manage big data. This allows for a chip-level defect analysis system that can recommend the relevant EDS test and identify the root cause in real time even with a single defective chip. The experimental results demonstrate that the proposed root cause search can reveal the hidden cause of a single defective chip by amplifying it with 90 vBADs, and system performance improves by a factor of 61.

CLUE: A CLUSTERING-BASED TEST REDUCTION APPROACH FOR SOFTWARE PRODUCT LINES

Nowadays, organizations have increasingly turned to software product lines (SPLs)/configurable systems to meet diverse user needs and market demands. SPLs offer configurability and scalability but pose a formidable challenge in testing. Assuring the quality of variants across numerous configurations demands innovative testing strategies that balance effectiveness and efficiency. To improve testing productivity, test reduction techniques have been widely employed in non-configurable code. However, test reduction for SPL systems remains mostly unexplored. This paper introduces CLUE, a novel test reduction approach to enhance testing productivity in SPLs. Our idea is that to effectively and efficiently reveal failures, tests should be distinctive and cover diverse behaviors of the system. Meanwhile, similar tests covering the same/similar behaviors of an SPL system do not need to be redundantly executed multiple times. CLUE clusters the system’s tests into distinctive groups containing similar tests. Within each cluster, tests are prioritized based on the number of feature interactions, a common root cause of defects in SPLs. CLUE continually selects and executes the top-prioritized test in each cluster until a bug is revealed or the allocated effort budget is exhausted. To evaluate CLUE, we conducted several experiments on a dataset of six widely-used SPL systems. The results show that CLUE enables developers to detect defects earlier, requiring up to 88% less effort than existing approaches. Additionally, using only 50% of the original test suites, CLUE can identify most of the bugs in the buggy SPL systems while maintaining fault localization performance.

MTG_CD: Multi-scale learnable transformation graph for fault classification and diagnosis in microservices

The rapid advancement of microservice architecture in the cloud has led to the necessity of effectively detecting, classifying, and diagnosing run failures in microservice applications. Due to the high dynamics of cloud environments and the complex dependencies between microservices, it is challenging to achieve robust real-time system fault identification. This paper proposes an interpretable fault diagnosis framework tailored for microservice architecture, namely Multi-scale Learnable Transformation Graph for Fault Classification and Diagnosis(MTG_CD). Firstly, we employ multi-scale neural transformation and graph structure adjacency matrix learning to enhance data diversity while extracting temporal-structural features from system monitoring metrics Secondly, a graph convolutional network (GCN) is utilized to fuse the extracted temporal-structural features in a multi-feature modeling approach, which helps to improve the accuracy of anomaly detection. To identify the root cause of system faults, we finally conduct a coarse-grained level diagnosis and exploration after obtaining the results of classifying the fault data. We evaluate the performance of MTG_CD on the microservice benchmark SockShop, demonstrating its superiority over several baseline methods in detecting CPU usage overhead, memory leak, and network delay faults. The average macro F1 score improves by 14.05%.

Identification of propagation path and root cause of faults based on generative adversarial networks in industrial systems

Fault diagnosis and detection (FDD) improves product quality, process safety, and profitability in modern industrial plants. Faults can propagate between different parts of plants due to connections between devices and control loops. As a result, control room operators may receive multiple alarms that distract their attention from addressing the main causes of abnormal situations. Therefore, investigating causal relationships between process variables is essential in determining the root cause of faults. This study proposes a new systematic approach for causality analysis to identify the propagation path and root cause of faults in industrial plants. The research's innovations are categorized into two main parts. Initially, the study utilizes machine learning algorithms to identify causal relationships using probabilistic concepts. Additionally, it introduces a novel architecture by merging the generative adversarial network (GAN) with the variational auto-encoder (VAE). The process variables are mapped to a latent space, where an adversarial generative network will investigate the causal relationship between variables. Finally, the Tennessee Eastman Process simulation and gas turbine industrial data are used to evaluate the proposed algorithm.

Fault Diagnosis of Wastewater Treatment Processes Based on CPSO-DKPCA

The wastewater treatment process (WWTP) is one of the most common links in chemical plants. However, the testing for diagnosing faults in wastewater treatment plants is expensive and time-consuming. Due to strong nonlinearity and variable autocorrelation, traditional WWTP diagnostic methods based on principal component analysis (PCA) can lead to low fault detection rates (FDR) or difficulty in determining the root cause of faults. In this paper, an improved dynamic kernel principal component analysis (DKPCA) and Granger causality (GC) analysis model that uses chaotic particle swarm optimization (CPSO) to detect WWTP and locate the root causes of faults is proposed. First, a kernel function is introduced to map a nonlinear matrix to a linear space. Then, the training data are extended through a time lag constant to solve the problem of nonlinear and variable autocorrelation in WWTP. Moreover, a novel fault candidate variables selection method, together with GC, is introduced to locate the root variables of the fault. The CPSO algorithm is employed to optimize DKPCA's kernel function parameters, enhancing the accuracy of fault monitoring and diagnosis models. Compared with traditional methods, the proposed method has a better fault detection rate, achieving 95.83% and 93.33% fault detection rates in simulated and real WWTP, respectively.

Solusi Berkelanjutan untuk Perbaikan Kualitas Innerbox di PT. BKI Surabaya

Surabaya is the center of a growing manufacturing industry that always strives to improve the quality of its products. One company that faces serious problems is PT. BKI, which faces a high rate of innerbox defects with potential negative impact on product quality. The study used Six Sigma methodology to identify and address the root cause of innerbox defects within two months. The results of the analysis show that the main causes of innerbox defects include structuring, material, management, environmental problems, as well as improper assembly methods. The proposed solution involves the use of duplex-type cardboard materials, improved assembly techniques, regular machine maintenance, and environmental control. The use of stripping bands instead of duct tape on innerbox packaging is an effective improvement. The implementation of control for two months showed an increase in sigma value from 2.94 to 2.83. The Six Sigma approach is proven to provide a systematic analytical foundation for significant improvement. The results help PT. BKI in improving quality, efficiency, and customer satisfaction

A New Hybrid Supervisory Control System for Cabinet-Type Firebox Furnaces

In this paper, an intelligent hybrid Industrial Control System (ICS) and a Supervisory Control System (SCS) are proposed to improve the efficiency, safety, availability, and control capabilities of industrial furnaces. The main components of ICS are process control systems and advanced control systems that consist of overheating protection and load control. New soft sensors are designed as a combination of Laguerre filters and an artificial neural network to estimate the surface temperature of the furnace’s tubes, which allows the protection system to adjust fuel flow rate via overriding commands. Model-based fault detection systems are developed to detect faults in the combustion system and fouling in the furnace tubes and prepare features for the supervisory system. The supervisory control system is responsible for interfering between different components, evaluating the situation, and decision making based on the unit status and process conditions. An intuitionistic fuzzy inference system is employed as the core of the supervisory controller to tolerate disturbance and faults by switching the control modes. Test studies using experimental data of the furnace indicate the capability of the proposed monitoring and control system to operate in various loading situations and recover the system from abnormal conditions. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In petrochemical industries, several reports have been issued about the load reduction of fired-heater furnaces imposed by combustion system faults and emergency shutdowns to carry out un-planned repairs due to fouling and wax-formation in tubes. Different activities such as detecting abnormal conditions, identifying faults, and enforcing corrective action can be performed by operators through manual actions. This paper is focused on designing a new supervisory control system (SCS) to be able to recover the fired heater furnace from abnormal conditions and keep running the plant. The SCS evaluates the condition of the unit by acquiring information from main variables, sensors, actuators, operating status of components and utilities, and operator commands. By identifying the root cause of faults, SCS makes decision on recognizing hazard degree, raising alarms, and applying automatic corrective actions.

Reducing Wall Shelf Product Defects Using the DMAIC Method in Furniture Industry Companies

The object of research is a place of business that produces household needs and wood-based industries. One of the products it produces is a minimalist wall shelf. In the period January –December 2021 minimalist wall shelf products are the products with the most defects. Manufacturing companies in this study have a product defect tolerance limit of 5% of the total production, but minimalist wall shelf products have a defect of 5.64%. This study aims to determine the types of defects that occur in wall shelf products, identify the causes of defects minimalist wall shelves and provide suggestions for improvements to reduce the number of future defects. In this research the method used is DMAIC. In the minimalist wall shelf production process, 4 types of defects were found. Based on the Pareto diagram, it was determined that the type of lacquer defect was not suitable to be the most dominant type of defect, namely 80.1%, it was also found that the DPMO value and sigma value on minimalist wall shelves were 14,100 and 3.69. Based on the analysis of the fishbone diagram method and Failure Mode and Effect Analysis (FMEA), it is known that the main root cause of product defects is a dirty, dusty and windy environment with an RPN (Risk Priority Number) value of 448. Analysis was carried out using the 5W + 1H approach to find improvements to overcome the problems that occur, obtained suggestions for improvements for the company, namely making work area picket schedules, making closed work areas, daily and monthly check sheetforms for maintenance of work tools, making varnish process work instructions, making training schedules for operators and submitting additional goods forms.

AI Empowered Net-RCA for 6G

6G is envisioned to offer higher data rate, improved reliability, ubiquitous AI services, and support massive scale of connected devices. As a consequence, 6G will be much more complex than its predecessors. The growth of the system scale and complexity as well as the coexistence with the legacy networks and the diversified service requirements will inevitably incur huge maintenance cost and efforts for future 6G networks. Network Root Cause Analysis (Net-RCA) plays a critical role in identifying root causes of network faults. In this article, we first give an introduction about the envisioned 6G networks. Next, we discuss the challenges and potential solutions of 6G network operation and management, and comprehensively survey existing RCA methods. Then we propose an artificial intelligence (AI)- empowered Net-RCA framework for 6G. Performance comparisons on both synthetic and real-world network data are carried out to demonstrate that the proposed method outperforms the existing method considerably.

Robust discriminant latent variable manifold learning for rotating machinery fault diagnosis

Fault diagnosis is an important technology for performing intelligent manufacturing. To simultaneously maintain high manufacturing quality and low failure rate for manufacturing systems, it is of great value to accurately locate the fault element, evaluate the fault severity and find the fault root cause. In order to effectively perform feature extraction for rotating machinery fault diagnosis, a robust discriminant latent variable manifold learning (RDLVML) algorithm is proposed in this paper. Specifically, RDLVML algorithm could effectively select the features of the high-dimensional fault data, and extract the low-dimensional fault features with better discrimination. Through the idea of discriminative learning, the accuracy of fault diagnosis was improved by increasing the aggregation of the same state feature samples and the difference between different classes of feature samples. A novel weighted neighborhood graph is proposed by constructing the q-Rényi kernel function, which could effectively suppress the interference of outliers and noise, and make the RDLVML algorithm more robust. Moreover, the fault diagnosis method of rotating machinery based on RDLVML is proposed in this paper. Through experimental verifications and comparisons with several classical feature selection algorithms, rotating machinery fault diagnosis could be more accurately performed by this method.

Pre-trained CNN-based TransUNet Model for Mixed-Type Defects in Wafer Maps

Classifying the patterns of defects in semiconductors is critical to finding the root cause of production defects. Especially as the concentration density and design complexity of semiconductor wafers increase, so do the size and severity of defects. The increased likelihood of mixed defects makes finding them more complex than traditional wafer defect detection methods. Manually inspecting wafers for defects is costly, creating a need for automated, artificial intelligence (AI)-based computer vision approaches. Previous research on defect analysis has several limitations, including low accuracy. To analyze mixed-type defects, existing research requires a separate model to be trained for each defect type, which is not scalable. In this paper, we propose a model for segmenting mixed defects by applying a pre-trained CNN-based TransUNet using N-pair contrastive loss. The proposed method allows you to extract an enhanced feature by repressing extraneous features and concentrating attention on the defects you want to discover. We evaluated the model on the Mixed-WM38 dataset with 38,015 images. The results of our experiments indicate that the suggested model performs better than previous works with an accuracy of 0.995 and an F1-Score of 0.995.

Automated program and software defect root cause analysis using machine learning techniques

For the automated root cause analysis (ARCA) method and simplified RCA technique, their empirical assessment is presented in this study. A focus group meeting is a foundation for the target problem identification in the ARCA technique. This is compared to earlier RCA methodologies which rely on problem sampling for target problem discovery and high beginning costs. In this research, we suggest a naïve Bayes based machine learning method for identifying the underlying causes of newly reported software issues, which will facilitate a quicker and more effective resolution of software bugs. The ARCA technique produced a large number of high-quality corrective actions while requiring a reasonable amount of effort. The strategy is an effective way to find new opportunities for process improvement and produce fresh process improvement ideas in contrast to the organization’s corporate practices. In addition it is simple to utilize. Ultimately, we compared the methodology with other machine learning classifiers including support vector machine and decision tree.

Machine Learning for Industry 4.0 [From the Guest Editors

The Fourth Industrial Revolution, also known as Industry 4.0, marks the technological shift from traditional manufacturing systems to smart cyberphysical systems. It leads to an improvement in overall productivity and a reduction in environmental impact and promotes sustainable economic development. Industry 4.0 has been driven by emerging technologies such as the Internet of Things (IoT), also called the Industrial Internet of Things; digital twins; artificial intelligence; cloud computing; and edge/fog computing <xref ref-type="bibr" rid="ref1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[1]</xref> , <xref ref-type="bibr" rid="ref2" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[2]</xref> , <xref ref-type="bibr" rid="ref3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[3]</xref> , <xref ref-type="bibr" rid="ref4" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[4]</xref> , <xref ref-type="bibr" rid="ref5" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[5]</xref> . It is a hot topic in both academia and industry. The implementation of IoT connects physical assets to cybernetworks and captures a significant amount of data. These data, often “big”, are then fed to AI-based mission-critical systems to perform production monitoring, quality inspection, fault root cause analysis, quality prediction, and process control. The proper adoption of relevant Industry 4.0 technologies should lead to significant efficiency improvements and cost reductions in various industrial sectors.

Multiscale Partial Symbolic Transfer Entropy for Time-Delay Root Cause Diagnosis in Nonstationary Industrial Processes

Nonstationary characteristics and lack of time delays analysis remain two large obstacles to diagnosing fault root cause in industrial processes. However, ignoring these two issues, many traditional methods tend to introduce some spurious or indirect causal relationships that interfere with root cause judgments. In this article, multiscale partial symbolic transfer entropy (MPSTE) is proposed to handle the above-mentioned problems. On the one hand, a multivariate multiscale filter module of MPSTE is designed to estimate the information transfer on multiple time scales and detect time delays while considering the indirect causality of the multivariate industrial system. On the other hand, to capture the causal information from the nonstationary processes, MPSTE introduces a segmental symbolization procedure and maps time series into finite rank vectors, overcoming the limitation of stationarity. Moreover, two causal criteria for MPSTE are constructed to identify the direction and significance level of causality correspondingly. MPSTE can thus provide sufficient information for causal relationships and exclude spurious causality. Finally, a causal diagram construction strategy with time delay based on MPSTE is established to diagnose the root cause and trace the fault propagation path. The results of experimental verification in a coal mill rig prove the effectiveness of the proposed strategy.

A Three-Step Framework for Multimodal Industrial Process Monitoring Based on DLAN, TSQTA, and FSBN

The process monitoring method for industrial production can technically achieve early warning of abnormal situations and help operators make timely and reliable response decisions. Because practical industrial processes have multimodal operating conditions, the data distributions of process variables are different. The different data distributions may cause the fault detection model to be invalid. In addition, the fault diagnosis model cannot find the correct root cause variable of system failure by only identifying abnormal variables. There are correlations between the trend states of the process variables. If we do not consider these correlations, this may result in an incorrect fault root cause. Therefore, multimodal industrial process monitoring is a tough issue. In this paper, we propose a three-step framework for multimodal industrial process monitoring. The framework aims for multimodal industrial processes to detect the faulty status timely and then find the correct root variable that causes the failure. We present deep local adaptive network (DLAN), two-stage qualitative trend analysis (TSQTA), and five-state Bayesian network (FSBN) to implement fault detection, identification, and diagnosis step by step. This framework can detect the system failure timely, identify abnormal variables, and find the root cause variable and the fault propagation path. The case studies on the Tennessee Eastman simulation and a practical chlorobenzene production process are provided to verify the effectiveness of the proposed framework in multimodal industrial process monitoring.

Improvement Productivity and Quality by Using Lean Six Sigma: A Case Study in Mechanical Manufacturing

Intelligent integrated production systems are always of interest to production planners. However, in order to deploy and improve from a mere processing fac- tory to a processing factory with an integrated intelligent production system, it requires a team of employees, engineers, and managers to always have a spirit of innovation, continuously improving existing semi-automatic equipment into automatic ones, aiming to move towards a smart factory. The result of this research is to reduce waste in the process of assembling mechanical products by applying the DMAIC process (Define, Measure, Analysis, Improve, Con- trol), lean six sigma tools, the test of variance, hypothesis testing and exper- imental design, IBM SPSS 2020, Matlab2019a and Minitab 18 software are used for data analysis and Solid work software is used to design and simulate mechanical parts. This study shows a systematic approach to analysis to find the root cause of defects in the product assembly process, a method of diag- nosing defective products as well as the application of charts to the analysis of waste products, and improving quality by applying basic quality manage- ment tools such as Pareto charts, fishbone diagrams, value stream mapping, man-machine chart, and failure tree analysis (FTA). Clearly identify the types of waste such as components sliding on top of each other without rust, and sur- face roughness of metal products that do not meet the standards. Experimental design models and statistical models, and statistical tests are applied to Lean six sigma’s DMAIC model in the process of analyzing the machining process. The results of analysis and process improvement have improved in a reduc- tion of scrap in the assembly line of mechanical products by 59.66% per year, an increase in assembly-line productivity by 7.8% per year, and a decrease in waste costs incurred by 59.66% per year. The application of the DMAIC cycle to improve the quality of the assembly line of mechanical products, in addition to reducing waste, also reduces the quality cost of the assembly line.