Mahalanobis Taguchi System (MTS) is developed by Genishi Taguchi as a part of his quality engineering methodology. The basic idea of Taguchi's quality engineering is looking for the way of effectiveness of analyzing multivariate system. In the MTS, with the standardized variables of healthy normal data, Mahalanobis Distance(MD) calculated and that can be discriminate between normal and abnormal objects. If this discrimination process is successful, next step is optimization which is try to reduce number of attributes by neglecting less effective attributes to MD. Orthogonal Array (OA) and Signal to Noise ratio (S/N) are used to evaluate the amount contribution of each attribute to the MD. Wisconsin Breast Cancer study, from machining learning repository at University of California at Irvine, used for examining the discriminant ability of MTS.

Process capability indices, Cp, Cpk, Cpm, and Cpmk can be used to evaluate single quality characteristic. Recently, many of the multivariate capability indices have been developed to assess a part which has several correlated quality characteristics. Principal component analysis(PCA) can transform the correlated quality characteristics into the newly independent factors. The existing formula calculates the Mutivariate Process Capability Index using the geometrical average. But, the result value notably differs from the number of principle component. In this paper, We proposed a new multivariate capability using PCA & RSS.

난지도 매립장 매립물의 특성을 조사하기 위해서 중력, 자력, LandSat TM 열적외선 밴드 자료, 매립장의 표면에서 측정한 침하량 자료 등을 다변량 통계분석기법을 응용하여 분석하였다. 분석에 이용한 자료들은 각기 상이한 깊이에 관한 정보를 제공하기 때문에 측정된 총 자력자료와 중력자료는 자극화변환된 자력이상과 매립장의 3차원 밀도분포로 각기 전환하였으며, 본 연구에서는 이 중 매립장의 상부층에 관한 정보를 이용하였다. 통계분석은 침하량 측정 지점들을 대상으로 수행하였으며, 이들 지점에서의 자극화변환 자력이상, 매립물의 밀도, LandSat TM 열적외선 밴드 값들은 내삽방법을 이용하여 구하였다. 자료분석에 사용한 다변량 통계분석 기법은 개체간의 기하학적인 거리를 이용하여 군집화하는 집락분석으로, 개체간의 거리 계산시 각 자료간의 상이한 측정단위가 주는 효과를 제거하기 위해서 사전에 표준화를 실시하였다. 군집화는 체계적 군집화 방법을 이용하여 수행하였다. 물리적 특성을 바탕으로 분류된 최적의 군집수는 수상도에서 나타난 결과에 따르면 총 6개의 군집으로 나타났다. 본 연구의 결과는 통합된 지구물리자료에 다변량 통계분석 기법을 적용함으로써 복합적 인 쓰레기 매립장의 특성 규명이 가능함을 시사한다.

고자리꽃파리 및 씨고자리꽃파리는 마늘, 양파 및 파 등 구근류 작물을 가해하는 중요한 농업해충이나, 그 형태적 유사성으로 인해 지금까지 학명적용에 있어서 학자간에 수많은 오류를 범하고 있다. 따라서 이들 2종에 대한 종간 및 각 종내의 지역별, 시기별, 기주식물별 등에 따른 형태적 정량형질의 다변량 해석을 실시하였다. 그 결과, 양종간에는 100%의 분리율로 명백히 구분되나, 2종 모두 지역별, 시기별 및 기주식물 등에 따른 종내 정량형질의 차이는 고자리꽃파리에서는 분리율이 모두 75% 이상으로 높았으며, 씨고자리꽃파리에서는 60% 이하로 낮게 나타났다.

국내 2개 선박회사로부터 입수한 1978년부터 1986년까지의 중요 기관고장에 관한 자료를 주성분분석법에 의하여 분석한 본 연구의 결과를 요약하면 다음과 같다. 1. 분석결과는 각성분의 인자부하량 크기에 따라 전기.자동제어기기, 보기류, 배관계, 과급기.냉동기류, 주기관계의 5그룹으로 나눌 수 있다. 2. 기기의 고장현상을 발열.소손, 고장원인은 부식.마모 및 오조작, 그리고 기기이상검출의 수단은 누설.혼탁 등이 매우 높은 비중을 차지하고 있다. 3. 주기관의 고장원인은 연료.윤활유 불량이 가장 큰 비중을 차지하며 주로 진동.이음에 의하여 이상을 검출하는 경우가 많다. 4. 전기.자동제어기기 고장은 피해도 비교적 적으며 기계계의 고장 피해는 큰 편이다. 5. 주기관 등은 기기이상의 확실한 사실에 의하여 이상이 검지되는 경향이 매우 강하며 보기류, 과급기.냉동기.에어콘 등은 운전원의 감각적 판단에 의하여 이상이 검지되는 경향이 강하다.

This study examines the behaviour of staple food price using Multivariate BEKK-GARCH Model. Understanding of staple food price behaviour is important for determining the unpredictability of staple food market and also for policy making. In this paper, we focus on the commodity prices of sugar, rice, soybean and wheat to examine the volatility behaviour of those commodities. The empirical results show that the own-volatility spillover are relatively significant for all food prices. The own-volatility spillover effect for sugar price is relatively large compared with the volatility spillover of other staple food commodities. The findings also highlight that the price volatility of wheat increases during food crisis more than it does when the condition is stable. Also, the own-volatility of rice and wheat in the period of the food crisis is significant and higher compared to the period before food crisis indicates that the past own-volatility effects during food crisis are relatively more difficult to predict because of the uncertainty and high price volatility. Policy recommendations that can be proposed based on the findings are: (1) a better trade agreement in food commodity trade, (2) Indonesia should not depend on wheat importation, and (3) early warning system to minimize food price volatility risks.

The adequacy of urban air quality monitoring networks in the largest metropolitan city, Seoul was evaluated using multivariate analysis for SO2, NO2, CO, PM10, and O3. Through cluster analysis for 5 air pollutants concentrations, existing monitoring stations are seen to be clustered mostly by geographical locations of the eight zones in Seoul. And the stations included in the same cluster are redundantly monitoring air pollutants exhibiting similar atmospheric behavior, thus it can be seen that they are being operated inefficiently. Because monitoring stations groups representing redudancy were different depending on measurement items and several pollutants are being measured at the same time in each air monitoring station, it is seemed to be not easy to integrate or transmigrate stations. But it may be proposed as follows : the redundant stations can be integrated or transmigrated based on ozone of which measures are increasing in recent years and alternatively the remaining pollutants other than the pollutant exhibiting similar atmospheric behavior with nearby station’s can be measured. So it is considered to be able to operate air quality monitoring networks effectively and economically in order to improve air quality.

Background : Curcuma longa L., in the family Zingiberaceae, is distributed in tropical and/or sub-tropical regions mainly in India and China. This species is commonly called turmeric, powder is used as medicinal herbs and/or flavor enhancer. It has been cultivated in southern region mainly Jindo. However, it might be possible to extend cultivation regions due to rise in average temperature. In order to select superior lines based on agronomic characteristics, we analyzed multivariate and estimated selection effects from C. longa germplasm. Methods and Results : The C. longa germplasm were cultivated in an experimental field located in Eumseong, NIHHS, RDA. The harvested roots were investigated in agronomic characteristics included in yield and then considered its relationship among the 9 germplasm by multivariate analysis method. Results from principal component analysis (PCoA) showed that it represented 70.00% and 80.44% accumulated explanation from four and five principal compounds (PC). PCoA was conducted from 9 agronomic characteristics and then correlation coefficient has been showed by analysis between each main component value and agronomic characteristics. Value of the first PC was 2.25, 24.96% explanation of total dispersion, plant height, number of rootlet and weight of rootlet were correlated with a somewhat higher level as 0.41, 0.43 and 0.52. Value of the fifth PC was 0.94, 10.43% explanation of total dispersion, the number of shoots was correlated with a higher level as 0.87. Selection effects with outstanding candidate lines including higher lines were estimated at 126.13% in yield. Conclusion : These data on multivariate based on agronomic characteristics will be give us invaluable breeding information by selection of superior lines.

In this study we established the high throughput screening system of high functional soybean cultivars using PLS modeling from FT-IR spectral data of soybean(Glycine max L) seeds. Crude extract of 20% methanol from soybean seed powders (153 lines) were used for FT-IR spectroscopy. Total fatty acid, carotenoids, flavonoids and phenolic compounds contents from soybean seed powders were analyzed using UV-spectrum and GC analysis respectively. PCA analysis showed that 153 soybean lines formed a single clusters with a few outlier. PC score 1 and 2 represented 39.5, 16.4% of total variation, respectively. And than showed change patten from the middle to outside for PCA plot. We conducted PLS regression analysis between FT-IR spectral data and fatty acids data. Palmitic acid showed the highest regression coefficient (R=0.78). This result implied that the content of palmitic acid could be predicted from FT-IR spectral data from soybean seed powders with relatively high fidelity. PLS modeling of total carotenoids also showed regression coefficient of 0.69. Regression coefficient of total flavnoids and phenolic compounds were 0.44, 0.39, respectively. At present, we are trying to confirm the accuracy of PLS prediction modeling using targeted metabolite analysis (GC-MS, LC-MS) from predicted soybean lines. To increase the accuracy of PLS modeling, we also trying to standardization of spectroscopy and spectral data processing. Furthermore we are going to develop PLS modeling from GC-MS, LC-MS data. The PLS prediction modeling established in this study could be applied for high throughput screening of other leguminous plant.

본 연구에서는 미계측유역에 대한 준분포형 강우-유출모형을 적용하기 위한 방법으로 두 개의 다변량 통계기법인 주성분분석과 계층적 군집분석을 연계한 매개변수 지역화 기법을 제안하였다. 109개 중권역 유역에 대해 7개 유역특성인자(유역면적, 평균표고, 평균경사, 산림면적비, 포화토양수분량, 포장용수량, 영구위조점)를 추출하였으며 주성분분석을 수행한 결과 제1, 2 성분이 전체자료의 82.11%를 설명하는 것으로 나타났다. 제1성분은 유역위치, 제2성분은 유

지역빈도해석을 통한 확률강우량 산정 결과는 수문학적으로 동질한 지역의 구분 결과에 따라 달라진다. 지역을 구분할 때에는 강우에 영향을 미치는 다양한 변수들이 사용될 수 있다. 변수의 유형과 개수가 지역 구분의 효율성을 좌우하기 때문에 활용 가능한 모든 변수들의 정보를 요약할 수 있는 변수들을 선택하는 것이 지역 구분의 효율성 면에서 유리하다고 할 수 있다. 이런 면에서 지역 구분의 효율성을 증대시킬 목적으로 다변량 분석 기법이 활용될 수 있다. 본 연구

This study was conducted to clarify the relation of the species of genus Sorbus in Korea based on multivariate analysis for the morphological characteristics and DNA polymorphism. Twenty-eight quantitative characters were assessed and analyzed by the principal component analysis and UPGMA cluster analysis. From the principal component analysis of 28 quantitative characters, three principal components (PC’s) explained the variation of inter-specific relations among the genus Sorbus. The first PC’s explained 58.95% of the variation with the Eigenvalue of 16.5, and the second and third PC’s showed the Eigenvalue of 8.3 and 3.1 and explained 88.74% and 100.0% of the variation, respectively. Especially, the first PC’s was related in order of the fruit width (FW) and length of terminal leaflet (LTL), petal length (PL), width of terminal leaflet (WTL), and diameter of winter bud (DWB). The second and third PC’s were involved in order of the No. of leaflet (NL), No. of fruit per fruiting lateral (NFL), length of upper rachis (LUR), and diameter of rachis (DR), No. of pistil (NP), respectively. Cluster analysis using them UPGMA method based on the principal components of four species of the genus Sorbus divided into two groups. Group Ⅰ comprises Sorbus commixta and S. sambucifolia var. pseudogracilis, and Group Ⅱ consists of S. amurensis and S. aucuparia. The pattern of DNA polymorphism of the 56 inter-simple sequence repeat (ISSR) markers revealed that different taxa shared different sets of bands, and DNA analysis is useful for taxonomic study on the genus Sorbus.

This study was conducted to evaluate characteristic of water quality in coastal sea of Incheon using a multivariate analysis. The analysis data in coastal sea of Incheon was aquired by the NFRDI data which was surveyed from March 1997 to November 2003. Eleven water quality parameters were determined on each survey. The results were summarized as follow ; Water quality in Incheon coastal sea could be explained up to 64.62% by three factors which were included in loading of fresh water and nutrients by the land(36.98%), seasonal variation(16.19%), and internal metabolism (11.24%). The results of time series analysis by factor score, in case of factor 1, station 1 influenced by Han river was shown to high factor score and station 3 located by outer sea was shown to low factor score. In case of factor 2, station 1 was appeared to high variation and station 3 was appeared to low variation. The result of cluster analysis by station was classified into three group that has different water quality characteristics. Especially, station 1 which affected by Han river and station 4 which affected by sewage treatment plant was appeared to considerable water quality characteristics against other station. In yearly cluster analysis, three group was classified and water quality in 2003 years due to high precipitation was different to another year. It could be suggested from these results that it is important to control discharge of fresh water by Han river and sewage treatment plant for water quality management of coastal sea of Incheon.

It is necessary that the basic concept of rural planning update from economics based on the production and sale into experience of natural resources and traditional culture. For the purpose of set up development direction for rural district, it is requisite to the multivariate analysis. In this study, the methods of the classification of rural village with existing data are studied, the results looking for applying to the making of principal viewpoint of the development. The analysis methods of classification are used the PCA, CA and combination of these, and making the revised method for localization of the rural district. In this study, we implement classification of regional pattern analysis for the planning of rural district in Chungbuk province.

This study was conducted to evaluate water quality in the Mankyung River using multivariate analysis. The analysis data which was surveyed from January 1996 to December 2002 in Mankyung river was aquired by the ministry of environment. Twelve water quality parameters were determined on each survey. The results were summarized as follow ; Water quality in the Mankyung River could be explained up to 74.90% by four factors which were included in loading of organic matter and nutrients by the tributaries(43.28%), seasonal variation(10.40%), loading of pathogenic bacteria by domestic sewage of Gapcheon (12.41%) and internal metabolism in river(8.81%). The result of cluster analysis by station was classified into three group that has different water quality characteristics. Especially, Iksan river was appeared to considerable water quality characteristics against other station. In monthly cluster analysis, three group was classified by seasonal characteristics. Also, in yearly cluster analysis, three group was classified. It is necessary to control the pollutant loadings by domestic sewage and livestock waste for water quality management of Mankyung river.

본 연구에서는 강우의 지역빈도분석에 필요한 수문학적 동질성을 갖는 지점강우의 권역화를 수행하였다. 이를 위해 전국에 걸친 기상청 산하의 60개 강우관측소에 대한 32개의 강우특성자료를 추출하였으며, 추출된 각 지점의 많은 강우자료들은 다변량 분석의 자료축약기법인 주성분분석과 그룹화 기법인 군집분석을 통하여 합리적이고 효율적으로 권역화되었다. 본 연구의 결과인 지점강우의 권역은 강우지역을 수문학적 동질성의 5개 권역과 3개의 기타지역으로 분류되었으며, 각