In factory automation, efforts are being made to increase productivity while maintaining high-quality products. In this study, a CNN network structure was designed to quickly and accurately recognize a cigarette located in the opposite direction or a cigarette with a loose end in an automated facility rotating at high speed for cigarette production. Tobacco inspection requires a simple network structure and fast processing time and performance. The proposed network has an excellent accuracy of 96.33% and a short processing time of 0.527 msec, showing excellent performance in learning time and performance compared to other CNN networks, confirming its practicality. In addition, it was confirmed that efficient learning is possible by increasing a small number of image data through a rotation conversion method.

This paper presents a log-transformed model-based performance analysis system for analyzing and improving manufacturing performance of the smart factory in the display business. Two years of data related to traditional manufacturing performance such as Cycle-time, WIP(Work-In-Process), and Throughput were investigated from the smart factory that producing the display for this research. We assessed manufacturing competitiveness based on how the operational level of automation affects improvements in manufacturing performances. We analyzed functional relationships between the indicators were derived using logtransformed regression analysis how the manufacturing performance indicators change according to the operational level of smart factory automation. As a result, we knew that the 170K production, which was planned capacity in the line design phase, achieved by running an automation level of only 59%. Based on this research, we suggest building an autopoietic optimize performance model to improving manufacturing competitiveness of smart manufacturing.

This paper considers the workforce assignment problem to minimize both the deviations of workloads assigned to workers and to maximize the total preference between each worker and each machine. Because of the high expense of technology education and the d

Given the plant layout, the number of workers, the maximum number of machines that a worker can handle, and the preferences between each worker and each machine, the problem to minimize the deviations of workloads assigned to workers and to maximize the total preference between each worker and each machine is considered. The number of workers are fixed (no part time workers) because of the high expense of technology education and the increase of current employees. Since the workforce assignment problem in this paper is in NP-class, a heuristic algorithm is presented in multiple-row plant layout during according to two types (slow and peak) of periods. The proposed algorithm is developed based on the combination of a mixed model scheduling, simulated annealing technique and graph theory. The solution generated satisfies the zone constraint (machines assigned to a worker are adjacently located). Computational results show that the presented algorithms can find a good solution quickly.