• Journal of the Korea Society of Computer and Information
• /
• v.7 no.1
• /
• pp.174-183
• /
• 2002

In this paper a control chart pattern recognition methodology based on the back propagation algorithm and Multi layer perceptron, a neural computing theory, is presented. This pattern recognition algorithm, suitable for real time statistical process control. evaluates observations routinely collected for control charting to determine whether a Pattern, such as a cycle. trend or shift, which is exists in the data. This approach is promising because of its flexible training and high speed computation with low-end workstation. The artificial neural network methodology is developed utilizing the delta learning rule, sigmoid activation function with two hidden layers. In a computer integrated manufacturing environment, the operator need not routinely monitor the control chart but, rather, can be alerted to patterns by a computer signal generated by the proposed system.

• Structural Engineering and Mechanics
• /
• v.71 no.3
• /
• pp.283-290
• /
• 2019

In global competition manufacturing companies have to produce modern, new constructions from advanced materials in order to increase competitiveness. The aim of my research was to develop a new composite cellular plate structure, which can be primarily used for structural elements of road, rail, water and air transport vehicles (e.g. vehicle bodies, ship floors). The new structure is novel and innovative, because all materials of the components of the newly developed structure are composites (laminated Carbon Fiber Reinforced Plastic (CFRP) deck plates with pultruded Glass Fiber Reinforced Plastic (GFRP) stiffeners), furthermore combines the characteristics of sandwich and cellular plate structures. The material of the structure is much more advantageous than traditional steel materials, due mainly to its low density, resulting in weight savings, causing lower fuel consumption and less environmental damage. In the study the optimal construction of a given geometry of a structural element of a road truck trailer body was defined by single- and multi-objective optimization (minimal cost and weight). During the single-objective optimization the Flexible Tolerance Optimization method, while during the multi-objective optimization the Particle Swarm Optimization method were used. Seven design constraints were considered: maximum deflection of the structure, buckling of the composite plates, buckling of the stiffeners, stress in the composite plates, stress in the stiffeners, eigenfrequency of the structure, size constraint for design variables. It was confirmed that the developed structure can be used principally as structural elements of transport vehicles and unit load devices (containers) and can be applied also in building construction.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.3
• /
• pp.148-154
• /
• 2021

We propose the AI Smart Factory Model for integrated management of production processes in this paper .It is an integrated platform system for the production of food packaging containers, consisting of a platform system for the main producer, one or more production partner platform systems, and one or more raw material partner platform systems while each subsystem of the three systems consists of an integrated storage server platform that can be expanded infinitely with flexible systems that can extend client PCs and main servers according to size and integrated management of overall raw materials and production-related information. The hardware collects production site information in real time by using various equipment such as PLCs, on-site PCs, barcode printers, and wireless APs at the production site. MES and e-SCM data are stored in the cloud database server to ensure security and high availability of data, and accumulated as big data. It was built based on the project focused on dissemination and diffusion of the smart factory construction, advancement, and easy maintenance system promoted by the Ministry of SMEs and Startups to enhance the competitiveness of small and medium-sized enterprises (SMEs) manufacturing sites while we plan to propose this model in the paper to state funding projects for SMEs.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.35 no.5
• /
• pp.509-515
• /
• 2022

As the 4th industrial revolution based on ICT is progressing in the manufacturing field, interest in building smart factories that can be flexible and customized according to customer demand is increasing. To this end, it is necessary to maximize the efficiency of factory by performing an automated process in real time through a network communication between engineers and equipment to be able to link the established IT system. It is also necessary to collect and store real-time data from heterogeneous facilities and to analyze and visualize a vast amount of data to utilize necessary information. Therefore, in this study, four types of controllers such as PLC, Arduino, Raspberry Pi, and embedded system, which are generally used to build a smart factory that can connect technologies such as artificial intelligence (AI), Internet of Things (IoT), and big data, are configured. This study was conducted for the development of a program that can collect and store data in real time to visualize and manage information. For communication verification by controller, data communication was implemented and verified with the data log in the program, and 3D monitoring was implemented and verified to check the process status such as planned quantity for each controller, actual quantity, production progress, operation rate, and defect rate.

• Design & Manufacturing
• /
• v.17 no.1
• /
• pp.7-12
• /
• 2023

Recently, with the emergence of the 4th industrial revolution, the demand for smart factories and factory automation is increasing. In this study, a tool life prediction program was developed to select optimal machining conditions using CNC milling equipment, which is widely used in flexible production and automation. The equipment used in the experiment was Hwacheon Machine Tool's 5-axis machining equipment, and the tool used was a 17F2R tool. For the machining path, the down-milling cutting method was selected and long-term machining was performed. The analysis standard for side wear on the tool was set at 0.1 to 0.2 mm, and tool life data and wear data were obtained in the cutting experiment. The program was created through the data obtained from the experiment, and a prediction rate of over 90% was secured when comparing the experimental data and the predicted data.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.27 no.4
• /
• pp.187-192
• /
• 2023

An elastic bearing must be strong against vertical loads and flexible against horizontal loads. However, due to the material characteristics of rubber, it may show variability due to the manufacturing process and environmental factors. If the value applied in the bridge design stage and the actual measured value have different values or if the performance during operation changes, the performance required in the design stage may not be achieved. In this paper, the seismic response of bridges was compared and analyzed by assuming a case where quality deviation occurs during construction compared to the design value for elastic bearings, which have not only always served as traditional bearings but also have had many applications in recent seismic reinforcement. The bearing's vertical stiffness and shear stiffness deviation were considered separately for the quality deviation. In order to investigate the seismic response, a time history analysis was performed using artificial seismic waves. The results confirmed that the change in the bearing's shear stiffness affects the natural period and response of the structure.

• Journal of Sensor Science and Technology
• /
• v.32 no.6
• /
• pp.386-390
• /
• 2023

Stiffness-engineered stretchable substrate technology has been widely used to produce stretchable displays, transistors, and integrated circuits because it is compatible with various flexible electronics technologies. However, the stiffness-engineering technology has never been applied to transistor-based stretchable strain sensors. In this study, we developed thin-film transistor-based strain sensors on stiffness-engineered stretchable substrates. We designed and fabricated strain-sensitive stretchable resistors capable of inducing changes in drain currents of transistors when subjected to stretching forces. The resistors and source electrodes of the transistors were connected in series to integrate the developed stretchable resistors with thin-film transistors on stretchable substrates by printing the resistors after fabricating transistors. The thin-film transistor-based stretchable strain sensors demonstrate feasibility as strain sensors operating under strains of 0%-5%. This strain range can be extended with further investigations. The proposed stiffness-engineering approach will expand the potential for the advancement and manufacturing of innovative stretchable strain sensors.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.1
• /
• pp.20-27
• /
• 2024

Determining the number of operators who set up the machines in a human-machine system is crucial for maximizing the benefits of automated production machines. A man-machine chart is an effective tool for identifying bottlenecks, improving process efficiency, and determining the optimal number of machines per operator. However, traditional man-machine charts are lacking in accounting for idle times, such as interruptions caused by other material handling equipment. We present an adjusted man-machine chart that determines the number of machines per operator, incorporating idleness as a penalty term. The adjusted man-machine chart efficiently deploys and schedules operators for the hole machining process to enhance productivity, where operators have various idle times, such as break times and waiting times by forklifts or trailers. Further, we conduct a simulation validation of traditional and proposed charts under various operational environments of operators' fixed and flexible break times. The simulation results indicate that the adjusted man-machine chart is better suited for real-world work environments and significantly improves productivity.

• Transactions of Materials Processing
• /
• v.33 no.2
• /
• pp.112-117
• /
• 2024

In this study, the friction and wear characteristics of Inconel 600 in the superplastic forming process of Ti6Al4V were evaluated through pin-on-disc tests. To achieve an efficient and systematic experimental design, the Taguchi method was employed. The wear track of the Inconel 600 pin showed scratches in the sliding contact direction, confirming that the wear mechanism is abrasive wear. Through sensitivity analysis such as ANOVA and Main effects, it was confirmed that both normal force and sliding distance have a significant impact on the wear. Changes in sliding velocity and distance did not affect the friction coefficient, which remained relatively constant at approximately 0.380. The wear prediction model for Inconel 600 in the superplastic forming of Ti6Al4V was constructed, which can be utilized as a guideline for the prediction and management of tool wear.

• Transactions of Materials Processing
• /
• v.33 no.2
• /
• pp.96-102
• /
• 2024

This study addresses evaluation of performance of hardening model for a titanium alloy (Ti6Al4V) based on the artificial neural network (ANN) regarding the strain rate and the temperature. Uniaxial compression tests were carried out at different strain rates from 0.001 /s to 10 /s and temperatures from 575 ℃ To 975 ℃. Using the experimental data, ANN models were trained and tested with different hyperparameters, such as size of hidden layer and optimizer. The input features were determined with the equivalent plastic strain, strain rate, and temperature while the output value was set to the equivalent stress. When the number of data is sufficient with a smooth tendency, both the Bayesian regulation (BR) and the Levenberg-Marquardt (LM) show good performance to predict the flow behavior. However, only BR algorithm shows a predictability when the number of data is insufficient. Furthermore, a proper size of the hidden layer must be confirmed to describe the behavior with the limited number of the data.