Demand for cosmetics with functionality and eco-friendliness has increased dramatically due to recent aging, well-being trends, and increased interest in beauty. Cosmetics production in 2014 was 8,970.4 billion won, an increase of about 50% compared to 6,014.6 billion won in 2010. In the midst of this, similar companies in intense competition are pursuing differentiated strategies and innovation activities to solve quality, price and delivery problems. In particular, cosmetics packaging work is getting more difficult due to the increasing bill of materials (BOM) and difficult assembly methods. Therefore, in this study, the following problems were identified and suggestions for the improvement of the packaging Many research laboratories such as biotechnology, chemistry, and pharmaceuticals, which are undergoing various studies, are equipped with ready-made laboratory safety equipments such as bio-safety workbenches, aseptic bases, and exhaust workbenches. However, most researchers are disadvantaged in using existing safety equipment. This is because existing safety equipment can not take into account all of the unique characteristics of the research. For this reason, researchers are demanding the development of customized safety equipment that is well suited to their research needs. process of Company C, which is facing difficult situation to respond to the customer 's delivery due to the 52 - hour work week. First, we used the stopwatch to find the difficulty process in the packaging process and show ways to improve it. Second, to improve the efficiency of line balancing in the packaging process, we integrate processes, improve work methods, and perform simple automation. As a result, the prepare loss for replacement was reduced by 1 minute from 5 minutes, resulting in a 23% increase in productivity from 112 ea./hour to 137ea./ hour per person. At this time, the LOB of the packaging process was improved from 70% to 82% by operating one more production line through one person per line, total 9 people saving.

요약; 최근에는 기계 학습, 특히 심층 학습에 많은 연구가 진행되고 있다. 구글, 페이스 북과 같은 대기업이 인공 지능과 기계 학습에 관심을 가지고 있기 때문에, 이러한 연구는 날마다 발전하고있다. 기계 학습은 의학, 번역 및 IT와 같은 다양한 산업에서 사용될 것으로 기대됩니다. 게임 부문은 기계학습 기술적용의 효과가 예상되는 영역 중 하나라고 간주됩니다. 본 논문에서는 MMORPG-Tera의 게임 콘텐츠에서 몬스터의 승패를 예측하는 신경망을 Tensorflow를 통해 설계하였다. 이 모델은 1 개의 입력 레이어, 2 개의 숨겨진 레이어 및 1 개의 출력 레이어를 가지고 있다. 입력 레이어에는 8 개의 노드가 있고 각 숨겨진 레이어에는 16 개의 노드가 있으며 출력 레이어에는 1 개의 노드가 있다. 더 나은 결과를 위해 우리는 그라디언트 디센트, 시그 모이드 (Sigmoid) 함수 및 Relu 함수 (Activate 함수)에 Adam을 사용한다. 준비된 데이터 세트의 마지막 부분은 테스 트 데이터 용으로 사용되고 나머지는 학습 모델 용으로 사용되었다. 이 모델은 5 ~ 10 % 오차 이내의 확률을 예측할 수 있다. 데이터 세트의 부족은 만족스럽지 않은 점으로 남아 있으며, 충분한 데이터가 수집되고 더 개선 된 모델이 준비되면 오류를 더 줄일 수 있다. 그리고 제안된 모델은 앞으로 다른 게임이나 스포츠 게임에도 적용될 것이다.

This thesis proposes a new social network simulator, which can be used for the social network analysis (SNA). It is composed of three modules; initialization, network evolution, and output generation. For the network evolution module, we suggest a modifie

Strawberry, Fragaria ananassa Duchesne, is one of the important horticultural crops cultivated in greenhouses. Tetranychus urticae is one of major strawberry pests, and Neoseiulus californicus and Phytoseiulus persimilis have been used as biological control agents for control of T. urticae. The interactions between T. urticae and N. californicus and between T. urticae and P. persimilis were investigated to compare their control efficiency for T. urticae on a spatially-structured strawberry leaf disc area at different temperatures and on different treatments. The experimental arena was an array of leaf discs (3 cm diameter) placed upside down on a water-saturated cotton pad in an aluminum pan (17.4 x 21.5 cm). Twenty leaf discs (4 x 5) were placed adjacent each other for allowing dispersal of T. urticae and its predatory mites. The temperature conditions were 20, 25, and 30°C and there were six different treatments. The overall population densities of T. urticae were influenced by temperature (20, 25, and 30°C) in N. californicus treatment. In the same temperature condition, P.persimilis was more effective than N. californicus to control T. uritcae. Two predator systems were better than one predator systems to suppress the population density of T. urticae at 25°C. Some results of this study could be used to understand the spatial association of T. urticae and its predatory mites in greenhouse crops and fields.

The two-spotted spider mite, Tetranychus urticae Koch, is one of major pests in greenhouse strawberry. Two predator mites, Neoseiulus californicus (McGregor) and Phytoseiulus persimilis (Athias-Henriot), have been widely used for control of T. urticae because they have good functional and numerical responses and searching behaviors. The study of single species releasing and combined releasing of two predatory mites, N. californicus and P. persimilis, was conducted on connected strawberry leaves. The experiments were run under laboratory conditions, 24±1oC, 50-65% RH, and a photoperiod of 16:8 (L:D) h. The excised leaf disk (diameter 3cm) of two strawberry varieties – Maehyang, Sulhyang– were placed upside down on a water-saturated cotton pad in an aluminum pan (width × length 17.4 × 21.5 cm). Twenty leaf disks were placed on each experimental set and the disks (width × length 4×5 cm.) were connected with each other for dispersing of T. urticae and its predatory mite. There were four different experiments – two strawberry varieties and two treatments (releasing single predatory mite, releasing two predator mites). The experiment sets were covered with plastic cage to protect from invading other insects and mites. All life stages of T. urticae and predatory mites were recorded until all mites were vanished. The data were transformed by ln (x+1). Repeated-measures analysis of variance was used to compare the temporal variation in the overall T. urticae and predatory mite density. The average number of T. urticae per leaf arena was significantly different among treatments in Sulhyang (Treatment, df=3, 196, F=17.86, P=0.0001; Time, df=6, 1176, F=47.76, P=0.0001; Time ×Treatment, df=18, 1176, F=22.06, P=0.0001) and in Maehyang (Treatment, df=3, 196, F=42.07, P=0.0001; Time, df=6, 1176, F=64.51, P=0.0001; Time x Treatment, df=18, 1176, F=24.19, P=0.0001). When N. californicus was introduced to P. persimilis system with diminishing prey, P. persimilis population increased more rapidly than N. alifornicus but P. persimilis was displaced by N. californicus. In single or combined releasing system, N. californicus persisted longer after prey depletion than P. persimilis. We examined population growth of P. persimilis and N. californicus in single and combined predatory mite released system with diminishing prey.

This study deals with the improvement of production process on a flow production system with the consideration of six sigma. We analyze the production process and survey the important factors of improvement of productivity. Using a six sigma, we find strategic point and suggest a reformation of production process. We applied a simulation technique to simulate the production line proposed by the result of the Six sigma. With the result of the simulation, this study analyzes the propriety of production line and proposes the alternatives of new production process.

This study deals with the improvement of production process on a flow production system with the consideration of six sigma. We analyze the production process and survey the important factors of improvement of productivity. Using a six sigma, we find strategic point and suggest a reformation of production process. We applied a simulation technique to simulate the production line proposed by the result of the Six sigma. With the result of the simulation, this study analyzes the propriety of production line and proposes the alternatives of new production process.

Petri Nets(PNs), as a graphical and mathematical tool, provide a uniform environment for modeling, formal analysis, and design of discrete event systems. The main objective of this paper is to model and analyze the manufacturing system using PNs as well as simulate the manufacturing system. The analysis method used in this paper is Embedded Markov Chain(EMC) method for Generalized Stochastic Petri nets(GSPN). The simulation tools used in this paper are Arena and Simnet V. 1.37. The results of three methods, EMC method, PNs simulation and Arena simulation, are obtained and compared.