• Journal of the Korea Safety Management & Science
• /
• v.2 no.2
• /
• pp.57-69
• /
• 2000

Beginning from July. 2000, the product liability of Korea is designed for reduction of customer loss by defective products. Therefore, most of company are supposed to be ready for taking care of safety and quality of products. ISO9001：2000 quality management system reform to emphasize continual improvement of the process with customer satisfaction. First of all, this paper start with introduction motive of the product liability and examine distinctive mark of the ISO9001：2000 quality management system. We consider the correlation between ISO9001：2000 quality management system and product liability and would like to propose product liability prevention by ISO9001：2000 quality management system with classification a defect style of product, the management guide and the law case.

• Journal of the Korean Society of Visualization
• /
• v.19 no.3
• /
• pp.63-68
• /
• 2021

In this study, data was augmentation through the SinGAN algorithm using small image data, and defects in rubber O-rings were detected using the random forest algorithm. Unlike the commonly used data augmentation image rotation method to solve the data imbalance problem, the data imbalance problem was solved by using the SinGAN algorithm. A study was conducted to distinguish between normal products and defective products of rubber o-ring by using the random forest algorithm. A total of 20,000 image date were divided into transit and testing datasets, and an accuracy result was obtained to distinguish 97.43% defects as a result of the test.

• Journal of the Korean Society of Systems Engineering
• /
• v.17 no.2
• /
• pp.1-8
• /
• 2021

This study proposes a systems engineering approach for the development of an advanced planning & scheduling (APS) system for a cosmetic case manufacturing factory. The APS system makes production plans and schedules based on the injection process, which consists of 27 plastic injection machines in parallel to control recommended inventory of products. The system uses machine operation/failure information and defective product/work-in-process tracking information to support intelligent scheduling. Furthermore, a genetic algorithm model is applied to handle the complexity of heuristic rules and machine/quality constraints in this process. As a result of the development, the recommended inventory compliance rate is improved by scheduling the 30-day production plan for 15 main products.

• Journal of the Korean Society for Precision Engineering
• /
• v.11 no.2
• /
• pp.189-196
• /
• 1994

Quality assurance in the process of manufacturing high strengh bolt sets used in the frictional joints of structures in one of important concern to users as well as to manufactures. In case of occurrences of either defective or low-quality products, even if they are quite rare, some systematic means of localizing the cause-characteristics and matching to corresponding production process is necessary. Control chart of torque factor is the primary indicatir in finding defectiveness of the products. Use of correlation diagrams ofnhardness of the bolt set presents in part a way of screening the cause. Retest data of the bolt set provide additional ideas of localizing the cause, for which theoretical background is presented in this regard. A process-characteristics matrix relating the causes of low quality to the corresponding process of manufacturing, which is of prime importance for the feedback control of production, is also proposed. Finally general features of control to assure quality of the set is described.

• Journal of the Korea Furniture Society
• /
• v.23 no.2
• /
• pp.110-121
• /
• 2012

With people's recent increasing interest in good design products, wood and metal products have gained great popularity. However, it was believed that it would be necessary to have a transformation to the manufacturing method based on digital data and equipments from existing analog-based manufacturing method, in order to meet consumers' demand. This study was aimed to seek for the possibility of mass-producing wood and metal products through the research on the type, usage and development conditions of digital data and the methods of utilizing digital equipments. As for research methods, the study analyzed the concepts and types of digital data through various internet and literature reviews and suggested perpetual calendar products as the final outcome of design development using computer data. Through this, the study summarized and organized actual design development processes by stage to provide basic data that could become the foundation of research on the design of wood and metal products using digital data. Through the outcome of this project, the following effects could be expected by developing wood and metal products through digital data. First, its accurate and precise process would help mass-produce complex forms of products and reduce their defective rate. Second, the compatible production of various types of digital equipments would lead to a cost reduction. Third, the diversity of design could be pursued by overcoming technical limitations. In order to satisfy the above expectation effects, such as realization of developing and producing various wood and metal products, there should be designers' creative experimental spirits, their active information exchange and cooperation with the companies concerned.

• Journal of Veterinary Clinics
• /
• v.39 no.1
• /
• pp.9-15
• /
• 2022

As more veterinary clinics become specialized with the growth of the companion animal market, an increasing number of veterinary clinics perform orthopedic surgery and use orthopedic products, some of which are defective and have side effects. Thus, the present study aimed to prepare fundamental data for the revision and development of manufacturing standards for these products in order to prevent their side effects. We conducted a survey targeting veterinary clinics as consumers and medical device companies as suppliers. Veterinary clinics were surveyed via offline and online methods; 320 clinics that offered orthopedic surgery and approximately 4,000 veterinary clinics that were registered in the Korean Veterinary Medical Association were targeted, and 153 veterinary clinics responded to the survey. The survey for medical device companies, was performed online, targeting 29 companies; 14 companies responded. The number of side effects of orthopedic products was higher in animal orthopedic products than in those for human use. Many consumers tended to suspect that side effects were caused by product defects. To resolve side effects after using orthopedic products, consumers mostly underwent reoperation. Meanwhile, some severe cases proceeded to legal disputes. Similarly, medical device companies, or the suppliers, responded that most side effects occurred in veterinary orthopedic products and that product defects and mistakes in use were the causes. As for most of the follow-up actions for side effects, these companies either reported the issue to those in charge or analyzed and resolved the issues themselves. Therefore, to develop quality products, suppliers should be provided with clear standards for the production, and information disclosure and a report system for side effects should be particularly established to gain consumers' trust regarding the safety of these products.

• Journal of Broadcast Engineering
• /
• v.21 no.3
• /
• pp.425-435
• /
• 2016

In this paper, we propose a classification model by analyzing raw material images recorded using a color CCD camera to automatically classify good and defective agricultural products such as rice, coffee, and green tea, and raw materials. The current classifying agricultural products mainly depends on visual selection by skilled laborers. However, classification ability may drop owing to repeated labor for a long period of time. To resolve the problems of existing human dependant commercial products, we propose a vision based automatic raw material classification combining mean shift clustering and stepwise region merging algorithm. In this paper, the image is divided into N cluster regions by applying the mean-shift clustering algorithm to the foreground map image. Second, the representative regions among the N cluster regions are selected and stepwise region-merging method is applied to integrate similar cluster regions by comparing both color and positional proximity to neighboring regions. The merged raw material objects thereby are expressed in a 2D color distribution of RG, GB, and BR. Third, a threshold is used to detect good and defective products based on color distribution ellipse for merged material objects. From the results of carrying out an experiment with diverse raw material images using the proposed method, less artificial manipulation by the user is required compared to existing clustering and commercial methods, and classification accuracy on raw materials is improved.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.125-140
• /
• 2018

As one of the processes in the manufacturing industry, quality inspection inspects the intermediate products or final products to separate the good-quality goods that meet the quality management standard and the defective goods that do not. The manual inspection of quality in a mass production system may result in low consistency and efficiency. Therefore, the quality inspection of mass-produced products involves automatic checking and classifying by the machines in many processes. Although there are many preceding studies on improving or optimizing the process using the data generated in the production process, there have been many constraints with regard to actual implementation due to the technical limitations of processing a large volume of data in real time. The recent research studies on big data have improved the data processing technology and enabled collecting, processing, and analyzing process data in real time. This paper aims to propose the process and details of applying big data for quality inspection and examine the applicability of the proposed method to the dairy industry. We review the previous studies and propose a big data analysis procedure that is applicable to the manufacturing sector. To assess the feasibility of the proposed method, we applied two methods to one of the quality inspection processes in the dairy industry: convolutional neural network and random forest. We collected, processed, and analyzed the images of caps and straws in real time, and then determined whether the products were defective or not. The result confirmed that there was a drastic increase in classification accuracy compared to the quality inspection performed in the past.

• Journal of the Korean Society of Systems Engineering
• /
• v.15 no.2
• /
• pp.57-65
• /
• 2019

The weapon system applies the rule that the system engineering procedure must be applied at the acquisition stage. Naval ships, which are one of the important naval weapon systems, take a long period of time to acquire, and the leadership must be commissioned. Therefore, in order to acquire a naval ship, high risk management is necessary, and it is necessary to promote business success through effective application of system engineering which is a scientific management method. However, in the case of naval ships, there are no specific procedures and methods for applying the system engineering. Therefore, research on application method of system engineering which can be easily used by person in charge is necessary. This should have a simple approach to ensure effective business performance by systematically refining and simplifying existing works. QFD (Quality Function Deployment) is a way to improve product satisfaction, impress customers, reduce defective products, reduce design errors, and reduce costs. This systematically develops these mutual relationships by converting the needs of customers into quality characteristics of products and defining them as the design quality of products considering the functions, quality, and process elements of parts. The purpose of this research is to present concrete methodology at the practical level using QFD in a way to ensure traceability of requirements which is an important element of system engineering.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.