• 산업경영시스템학회지
• /
• 제44권4호
• /
• pp.12-22
• /
• 2021

This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on k-fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, F₁-score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.

• Design & Manufacturing
• /
• 제11권1호
• /
• pp.30-33
• /
• 2017

In this study, the mold technology for manufacturing of porous implant was investigated. Firstly, we considered the concept of insert molding technology with 3D printing of porous inert part. The part on implant was designed in the end region of the implant. And then main implant bodies were manufactured using conventional machining method. The other porous parts were designed and optimized with molding simulation. As the feature size of porous implant was so small that perfect feature of it using 3D printing technology could not be obtained. So, we proposed another scheme for manufacturing of the porous implant in the replace of the former approach. Polymer mold cores with 3D printing technology were considered. The effects of addictive manufacturing process parameters on the properties of mechanical and dimensional accuracy were investigated. Direct 3D printed polymer mold cores were designed and manufactured under the simulation of thermal and molding analysis. It was shown that 3D printed mold core with polymer could be adapted to the injection molding for porous implant.

• 한국정밀공학회지
• /
• 제20권12호
• /
• pp.25-33
• /
• 2003

A burr has been defined as undesirable projection of material formed as a result of plastic flow from a cutting or shearing operation. It is unavoidable in all kinds of machining operation. As a result, burr makes troubles on manufacturing process due to deburring cost, quality of products and productivity. In this study, the primary interest is about exit burr. The burr formation mechanism in each type of burr is classified. Data bases are developed to predict burr formation result. In the milling operation, we develop an algorithm to analyze the burr formation mechanism by the geometrical analysis on the multi featured workpiece with multi cutting path. The algorithm includes three steps, i. e., the feature identification, the cutting condition identification, and the analysis on exit burr formation. We can predict which portion of workpiece would have the exit burr in advance so that we can manage to find a way to minimize the exit burr formation in an actual cutting. Also, this algorithm can be implemented in a commercial CAM package so that we can simulate the NC code to review the burr formation in advance.

• 한국전자거래학회지
• /
• 제8권4호
• /
• pp.53-68
• /
• 2003

Currently, it is possible to buy almost anything from books(Amazon.com) to airplane tickets(Travelocity.com) using the world wide web. The purpose of this research is to develop a "clean interface" between design and fabrication facilities for the production of custom machined parts through Internet. The current mechanism for production of prototype parts that can be fabricated using standard machine tools like milling machines, requires a process of part description preparation, bidding, contract award, and finally fabrication and delivery of the part. This is a substantially more complex process than buying a book or airplane ticket. In this paper, we try to define the ambiguous part description using XML based data exchange format and to enable e-commerce in this field. The research accomplishments are summarized: 1. Creation of a new format for data exchange of machined prototype parts, 2. Development of a prototype system to illustrate how the XML data can be effectively used to conduct e-Commerce for custom machined parts, 3. Testing of the methodology with a number of parts.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2004년도 학술대회지
• /
• pp.68-72
• /
• 2004

A practical method to investigate the user interface of web based Flexible Manufacturing System(FMS) on the internet environment is established. Because the industrial FMS controller requires a lot of gadget, such as switch, dial, button, etc., for actual work-site flexible operation sufficiently, the user interface of the controller is significantly complex. The support for operational convenience of FMS controller can increase productivity and efficiency of the user, operational personnel of FMS. While most FMS provide their application programming interface(API) and graphical user interface(GUI) with adequate mechanism itself when used in stand alone, there is increasing demand for FMS that can operate with the intuitional user interface find virtual reality(V/R) environment. This thesis considers the intuitional user interface of Web-based FMS first, and from this, goes a step further, improves as virtual reality environment of FMS on the internet environments by using the feature based modeling technique approach and cartoon rendering. The feature-based modeling technique approach is applied to FMS line which is consist of facilities such as machining center, CNC lathe, autonomous guided vehicle, rail guided vehicle, and various controllers. In this study, the FMS established the intuitional user interface is able to obtain not only the operational convenience but also the enough productivity and significant efficiency.

• 대한산업공학회지
• /
• 제20권4호
• /
• pp.153-180
• /
• 1994

The assemblability is determined by the structure of product and the relationship between composing parts and machining parts. In this paper, the bolt was divided into bolt-head, -shaft, -thread and -end. For higher assemblability in bolting process, it was analyzed the geometric and technological characteristics of bolts were analysed regarding pre- and in-assembly process. And this paper presents the knowledge-based expert system to assist for designer in the processor of designing bolt for easier assembly. The developed expert system for supporting bolt design assemblability which is named as BDFA SYSTEM consists of two system such as "BOLT DESIGN SYSTEM" which provide feasible assembly bolt design to designer and "EVALUATION SYSTEM" which provide assembly evaluation to alternative of bolt design.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1993년도 추계학술대회 논문집
• /
• pp.70-75
• /
• 1993

Chatter vibration is an unwanted phenomenon in metal cutting and it always affects surface finish, tool life machine life and the productivity of machining process. The In-process monitoring & control of chatter vibration is necessarily required to automation system. In this study, we constructed the multi-sensing system using Tool Dynamometer,Accelerometer and AE(Acoustic Emission) sensor for the credible detection of chatter vibration. And a new approach using a neural network to process the features of multi-sensor for the recognition of chatter vibration in turning operation is proposed. With the back propagation training process, the neural network memorize and classify the feature difference of multi-sensor signals.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 춘계학술대회 논문집
• /
• pp.211-214
• /
• 2002

The detection of tool wear is very important in an automated manufacturing system. This paper presents a tool condition monitoring system based on the wavelet transform analysis of the AC servo motor current in a milling process. The current measurement is relatively simple and does not affect machining operations. The discrete wavelet transform was used to decompose the current of a spindle AC servo motor in the time and frequency domain. The feature vectors were extracted from the decomposed signals and compared to clarity normal and wear conditions. The results show the feasibility of the wavelet transform analysis for the tool condition monitoring.

• 한국정밀공학회지
• /
• 제17권2호
• /
• pp.73-79
• /
• 2000

The part-surface of mold or stamping-dies consists of a compound surface which consists of lots of composite surfaces, and may have various types of feature shapes including convex sharp edge (CSE). Those CSE features should be considered with care in machining the surface, which necessitates extraction of CSE curves on a compound surface. This work can be done rather easily for a solid model which has a complete topology information. In case of the compound surface without topology information, however, such CSE curves must be gathered through some geometrical calculations paying much computation time. In the paper, extracting CSE curves by the construction of a CSE region-map which can reduce time, and detecting various common edge types are presented.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2002년도 춘계학술대회 논문집
• /
• pp.86-91
• /
• 2002

The production part requiring multiple processes such as piecing, blanking and notching, are performed with a high production rates in progressive die. In order to prevent the dejects of process result, the optimum of strip process layout design, die design, die making, and tryout with inspection etc. are needed. According to these factors of die development process, they required theory and practice of metal working process and its phenomena, die structure, machining conditions for die making, die materials, heat treatment of die camponents, know-how and so on. In this study, we designed and analyzed die camponents also simulated the strip process layout of multiple stage drawing by DEFORM. Especially the result of tryout and its analysis become to the feature of this study.