Crystal structure of DNA photolyase from Escherichia coli

박희원 Univ. of Texas Southwestern Medical Center at Dall 1995 해외박사

Photolyase는 자외선에 의해 만들어진 DAN damage, the cis, syn-cyclobuta ne pyrimidine dimer(pyr=Pyr)를 고치는 효과입니다. 이 효소 작용은 blue l ight를 흡수함으로서 시작되어서 long range energy transduction, single el ectron transfer, and enzyme catalysis by radical mechanism 등을 포함하고 있다. 이번 실험에 쓰여진 효소는 flavin이 radical 상태이고 folate는 mono glutamate 상태이다. 두 가지 종류의 crystal이 만들어 졌는데 두 종류 모두 symmetry는 space group P1 이었다. 첫번째 crystal의 unit cell 크기는 a=62. 1Å, b=72. 2Å, c=97. 3Å, α=97. 4°, β=108. 3°, γ=90. 0° 두번째 crystal의 unit cell 크기는 a=62. 6Å, b=72. 2Å, c=58. 5Å, α=99. 1°, β=101. 5°, γ=72. 0° 각각의 unit cell은 두개의 효소로 이루어져 있다. 원자 구조를 풀기 위하 여 여러 방법이 혼용되었는데 그 방법들은 다음과 같다: multiple isomorphou s replacement, solvent flattening, non-crystallographic symmetry averagi ng, and crystallographic refinement. 최종적인 구조는 2. 3Å resolution 까지 계산되었고 R-value는 0. 176 이었다. 원자 구조를 자세하게 살펴보면 α/β domain과 (dinucleotide binding domain을 닮음) helical domain으로 나누어진다. Folate가 (the light harvesting cofactor) 두 domain 사이에 박혀 있다. Energy가 folate로 부터 flavin으로 (the catalytic cofactor) 전해지는데 거리가 16. 8Å 정도 떨어져 있다. Flavin은 helical domain에 감 싸여 있고 이 domain에 있는 hole를 통해서 damaged DNA 와 접촉한다. Hole 의 크기와 모양이 pyr=Pyr과 잘 맞아떨어지므로 pyr=Pyr가 DNA helix 에서 빠져나와서 hole안으로 들어가는 "flip out" mechanism를 암시한다. 바 로 이때 electron이 쉽게 flavin으로 부터 pyr=Pyr까지 아주 짧은 거리에서 전해지도록 하는 것이 가능해 진다. 마지막으로, flash light을 효소에 조사 함으로써 electron이 Trp 306로 부터 radical 상태에 있는 flavin에 전해지는 기전에 관해 원자구조와 연계해서 살펴보면 여러 개의 pathway를 통해서 elec tron이 전해질 수 있음을 안다. Photolyase repairs ultraviolet damage to DNA by splitting the cyclobutane ring of.the major UV photoproduct, the cis,syn - cyclobutane pyrimidine dimer (Pyr<>Pyr). The reaction is initiated by blue light and involves long range energy transduction, single electron transfer, and enzyme catalysis by radical mechanism. Photolyase used in this study had the flavin in the neutral radical form and the folate in the monoglutamate form. Two crystal forms of the enzyme were obtained; both have symmetry of space group Pl. Form I has the unit cell dimensions Diffraction from this form extends beyond 3A resolution, but the crystals are radiation-sensitive and difficult to reproduce. Form II has the unit cell dimensions a the unit cell contains two molecules. The crystal structure was determined by a combination of multiple isomorphous replacement, solvent flattening, non-crystallographic symmetry averaging, and crystallographic refinement. The refined atomic model has an R-value of 0.176 at 2.3A resolution. The polypeptide chain of 471 amino acids is folded into an amino-terminal cr./p domain resembling dinucleotide binding domains, and a helical domain. A loop of 72 residues connects the domains. The light harvesting co-factor 5,10-methenyltetrahydrofolate is attached in a cleft between the two domains. Energy transfer from MTHF to the catalytic cofactor, flavin adenine dinucleotide (FAD) occurs over a distance of 168A. The FAD cofactor adopts a U-shaped conformation between two helix clusters in the center of the helical domain and is accessible through a hole in the surface of this domain. Dimensions and polarity of the hole match those of a PyroPyr dinucleotide, suggesting that the PyroPyr lips out of the helix to fit into this hole, and that electron transfer between the flavin and the PyroPyr occurs over van der Waals contact distance. During purification, the flavin is oxidized to the catalytically inert form, FADHS This form can be reactivated by irradiation with white light which causes electron transfer from Trp306 to the excited FADHO. On the basis of the structure, several possible pathways can be proposed for electron transfer from Trp306 to FAD.

Comprehensibility and availability of input to adult learners of L2, with special reference to input hypothesis and comprehensible input

한문섭 Univ. of Texas at Austin 1991 해외박사

본 연구는 어떻게 그리고 어떠한 환경에서 학습자가 이해가능 입력자료를 진행시키는가를 알아보는 것을 목적으로 한다. 이론적 근거는 Krashen의 입 력가설과 이해가능 입력자료이다. 입력가설에 의하면 학습자는 이해가능 입 력자료 (i + 1)를 이해함으로써 언어를 습득하게 된다. 입력가설은 많은 비 판을 받았음에도 불구하고, 이해가능 입력자료라고 하는 Krashen의 생각은 오늘날 폭넓게 받아들여지고 있으며, 그의 논지는 제 2 언어습득분야에 많은 영향을 미치게 되었다. 그런데, 이 가설의 문제점중 하나는 이 가설은 직접적으로 실험이 가능하지 않다는 것인데, 그 이유는 이 가설의 설명에 사용된 대부분의 용어들이 실험 에 직접 사용될 수 없기 때문이다. 그러므로, 이 가설의 내용 전반에 걸친 실험보다는, 이 가설의 주 내용에 해당하는 이해가능입력의 부분적 개념에 대 해서만 실제 실험이 이루어져 왔고, 본 논문 또한 이런 방식의 부분적 실험만 실시하였다. 본 논문의 실험에 사용된 입력자료는 학습자의 학습환경에 따라 세가지로 분류하였다. 즉, interaction input,classroom input, non-interaction inpu t의 세 종류가 실제 실험에 사용되었다. 외국어학습 환경에서는 자연적 환경 과 교실 환경에서 받아들이는 이해가능 입력자료의 양과 질이 크게 제한되기 때문에, non-interaction input이 습득에 직접 도움이 될 수 있는 이해가능 입력자료를 학습자에게 제공할 것으로 예측되었다. 본 연구에서 연구과제로 제시된 것은, 학습자가 이해가능 입력자료를 충분 히 받아들이기 위해서는, 원어민과의 접촉이 유일한 수단인가 하는 것이다. 이러한 연구과제를 수행하기 위해서 다음과 같은 두 가지의 실험가설이 세워 졌다. 가설 1 : 학습된 지식에 의한 언어능력과 습득된 지식에 의한 언어 능력에 는 차이가 없다. 가설 2 : 원어민과의 접촉이 없으면 이해가능 입력자료를 제대로 받아들이 지 못할 것이다. The present study attempts to find how and in what environment learners process comprehensible input. The theoretical focus is on Krashen's input hypothesis and comprehensible input. The input hypothesis argues that learners acquire a language simply by understanding comprehensible input (i + 1). Even though the input hypotheSiS has received much criticism, Krashen'S idea of comprehensjble input is widely accepted by many researchers these days, and his argumentS have brought many insightfUl ideas to second language acquisition theories. One of the major limitations of this hypothesis is that it is not directly researchable, due to certain terms that can not be operationalized. Therefore, instead of testing the whole concept of the hypothesis, research was conducted on some aspects of the comprehensible input, which constitutes a major portion of input hypothesis. Three kinds of input based on the learner's environment have been postulated: interaction (natural) input , classroom input, and non-interaction input. As the foreign language environment provides limited quality and quantity of comprehensible input from natural and classroom environments, it was hypothesjzed that non-interaction input may provide learners with comprehensjble input that leads to acquisition. The central question investigated in this study is whether interaction with L2 native speakers is the only way that learners can receive enough comprehensjble input. Two hypotheses were put forth to answer the three research questions addressed. Forty Korean graduate students at the Unjversity of Texas at Austin were selected as subjects. Two kinds of tests, Stress Elicitation Test (SET) and Direct Stress Test (DST), as well as a Language Background Questionnaire (LBQ) were prepared to test the hypotheses. The findings indicate that, while interaction input is best for acquisjtion, as proposed by many researchers, non-interaction input is more important than classroom input at least in the foreign language learning setting. Additional findings are: (1) formally learned competence cannot become an acquired competence with comprehensible input, (2) learners can make input comprehensible to themselves without direct interaction wih L2 native speakers.

Solution-based Chemical Synthesis of Electrode Materials for Electrochemical Power Sources

정연욱 Univ. of Texas at Austin 2003 해외박사

The popularity of portable electronic devices and the desire for clean-air vehicles have created enormous interest in electrochemical power sources. Lithium-ion batteries offering higher energy density compared to other rechargeable battery systems are becoming the choice of a power source for portables. On the other hand, electrochemical supercapacitors offering higher power density compared to batteries are appealing for hybrid electric vehicles. However, both the lithium-ion and supercapacitor technologies are hampered by the high cost and toxicity of the currently used electrode materials. This dissertation explores alternate low cost materials for lithium-ion batteries and supercapacitors by employing novel solution-based chemical synthesis procedures. Manganese oxides are attractive alternates for lithium-ion cells as Mn is inexpensive and environmentally benign. Several Na_(x)MnO_(2+δ) oxides crystallizing in different structures have been synthesized in this study by reducing sodium permanganate with sodium iodide in aqueous medium followed by firing the reduction product. While the hexagonal Na_(0.7)MnO_(2+δ) (d □ 0.3) transforms to a spinel-like phase during ion exchange with lithium salts, the layered and tunnel Na_(0.5)MnO_(2+δ) are quite stable to ion-exchange reactions. The ion-exchanged layered and tunnel Na_(0.5-x)Li_(x)MnO_(2+δ) exhibit initial capacities of, respectively, 225 and 170 mA/g. While it is difficult to maintain a high capacity with good cyclability with the tunnel structure, the layered material is found to exhibit good cyclability. Amorphous RuO_(2)·xH_(2)O has been shown in the literature to exhibit a high capacitance of 720 F/g in electrochemical redox capacitors. With an objective to lower the cost per F capacitance, (i) substitutions of low cost Cr and W for Ru, (ii) coating of RuO_(2)·xH_(2)O on low cost oxides, and (iii) other low cost transition metal oxides and sulfides in various electrolytes have been pursued in this study. Ru_(1-x)Cr_(x)O_(2)·xH_(2)O, Ru_(1-x)W_(x)O_(2)·xH_(2)O and RuO_(2)·xH_(2)O coated on Na_(0.37)WO_(3)·xH_(2)O show capacitance values of, respectively, 840, 560 and 450 F/g in acid electrolyte, which are higher than that expected based on the rule of mixture. Amorphous MnO_(2)·xH_(2)O exhibits 250 F/g in NaCl electrolyte. Amorphous and crystalline VO_(2)·xH_(2)O, Fe_(2)O_(3)·xH_(2)O, MoO_(2)·xH_(2)O, NiS_(y) and CoS_(y) have also been investigated in various electrolytes, but their capacitance values are too low to be commercially viable.

Man-Machine Balanced Control for Automation of Infrastruckture Crack Sealing

김영석 Univ. of Texas at Austin 1997 해외박사

建築 및 土木構造物들에 있어 老朽化의 現象은 表面龜裂(Surface Cracking)의 形態로 가장 먼저 視覺化되어진다. 大槪의 境遇에 있어 構造物 表面에 發生되어진 龜裂들이 建物의 崩壞와 같은 構造物의 構造的 動向(Structural Action)이나 壽命(Durability)에 있어 直接的 인 影響을 주는 것은 아니지만 建築 및 土木構造物의 施工과 維持補修期間동안에 發生되어진 龜裂들이 該當 構造物들의 品質性 및 商品性을 低下시킬수 있으며 또한 時宜適切하고 效果的인 龜裂補修가 構造物의 壽命(Service Life)과 品質을 向上시킬 수 있다는 點에 있어서 이러한 龜裂들에 對한 補修는 專門家들의 安全診斷을 通하여 반드시 施行되어져야만 할 것이다. 이 論文은 建築 및 土木構造物에 發生되어진 龜裂들을 效果的이고 能率的으로 補修하기 爲한 하나의 革新的인 自動化 모델을 提示하고 있다. 開發되어진 自動化 모델을 構成하는 主要內容으로서 構造物들에 있어 補修되어질 龜裂들의 正確한 位置 探知 및 龜裂의 自動化 맵핑 過程과 探知되어진 龜裂들의 能率的 補修를 爲한 最短 經路 計劃 等이 이 論文에 있어 詳細히 紹介되어진다. 또한 龜裂補修 自動化 시스템들이 開發에 있어 共通的으로 要求되어지는 主要 硏究 課題들과 디자인 要素들이 이 論文에서 提示되어지고 論하여 진다. 이러한 自動化 모델은 텍사스 州立大學에 依해 開發되어진 道路 龜裂補修 自動化 시스템에 適用되어 왔으며, 모델의 效率成 및 實用性 그리고 生産性 考察을 通한 龜裂補修 自動化 工法의 優秀性을 立證하기 爲한 廣範圍한 實驗과 分析들이 持續되어져 왔다. 또한 道路 以外의 建築 및 土木構造物들에 對한 모델의 一般性 및 效率性을 立證하기 爲하여 開發되어진 自動化 모델과 多樣한 表面龜裂 이미지(Image) 들을 利用한 實驗結果들이 이 論文에서 提示되어지고 論하여 진다. 끝으로 建築 및 土木構造物들에 있어 龜裂補修를 自動化하는 것이 費用節減의 效果뿐만 아니라 安全性(Safety), 生産性(Productivity), 그리고 品質性(Quality)에 對한 向上을 가져올 수 있을 것이며 提示되어진 龜裂補修 自動化 모델이 道路 以外의 빌딩, 高層아파트, 上下水道管 및 가스管, 貯藏用 탱크, 터널 等과 같은 거의 大部分 構造物들에 있어 發生되어지는 龜裂補修를 爲하여 效果的으로 應用되어질 수 있을 것으로 思料되어진다. 그리고 앞으로 最短時日內에 이 硏究의 結果 道路 以外의 모든 龜裂補修 自動化 시스템들의 開發에 있어 보다 進步된 尖端 技術과 裝備가 動員되는 技能 發展에 拍車를 加할 것으로 期待되어진다. The bacteriophage λO protein localizes the initiation of replication to a unique sequence, ori λ through specific protein-DNA and protein-protein interactions. Conformational changes at ori λ introduced by O protein binding have been reported and their roles have been implicated in the initiation of λ DNA replication. In our studies, we focused on detailed structural basis of the formation of the O protein-ori λ complex(O-some). This work was divided into five Chapters including a general introduction given in Chapter 1, a brief review of genetic, biochemical, and structural data for understanding the mechanism of the initiation of λ DNA replication. As shown in Chapter 2, we found that the O protein exists as a dimer and demonstrated that the active DNA binding species is also a dimer. Dimerization and sequence-specific DNA recognition are specifically mediated through the amino-terminal half of O(O1-162(아래첨자 1-162입력불가)). The binding affinity of O for a single copy of its 19 bp recognition sequence was 2-3 nM. We also found that the O-some is composed of 4 dimers of O and ori λ DNA, which contains four 19 bp direct repeat recognition sites, i, e., a dimer of O is bound to each repeat(iteron). Moreover, we found that only the amino-terminal DNA binding domain is required for formation of the O-some. To investigate the structural basis for the unique properties of O protein, we generated a number of carboxy-terminal and internal and internal deletion mutants of O. Experiments with purified mutant proteins, as shown in Chapter 3, indicated that (ⅰ) the deletion mutant retaining amino acid residues 19-110 is the smallest O protein species that can both bind to DNA and form a dimer, (ⅱ) the affinities of all mutant proteins for a single iteron are almost the same, ranging from 2 to 4 nM; (ⅲ) the portion of O that is responsible for dimerization is located between amino acid residues 19 and 85; (ⅳ) the carboxy-terminal domain (O 156-299(아래첨자 156-299입력불가)) is a monomeric species that does not recognize specific DNA sequences but instead, bind non-specifically to duplex DNA; (ⅴ) the linker joining the two structural domains is not required for O function, but its coding sequence of DNA contains several recognition sites for O protein (ori λ); and (ⅵ) a deletion! m! utant missing the amino-terminal portion of the carboxyl-terminal domain is still comparably active in the in vitro λdv replication assay. In Chapter 4, the structural basis of the O protein-DNA complex was studied in detail. Hydroxy radical footprinting was employed to obtain the high resolution structural information about the contacts between the protein and the sugar-phosphate backbone of DNA. The missing nucleoside experiment allowed us to identify energetically important base moieties that may be in contact with bound O protein. Quantitation of the extent of O-mediated DNA bending indicated that O induces a relatively sharp bend in an individual recognition sequence of 85。 ±5。 . Measurement of the O-induced topological change indicated that a region of DNA or specifically ori λis wrapped around the O protein core in a left-handed fashion with a linking number change of 0.7±0.1 turn. In Chapter 5, we present direct evidence that the O protein also has the capacity to interact with single-stranded DNA, the first such interaction discovered among prokaryotic origin-binding proteins. The implication of this dual DNA binding specificity of O for the formation of the unwound structure at the A/T-rich region of ori λ is dis cussed. The addition of the λP-DnaB comple x to the O-some produces a new nucleoprotein species with a super-shift in migration. The presence of P and DnaB reduces significantly the amount of O required for binding to single-stranded DNA. Based on these results, we propose a detailed model for sequential structural changes in ori λ as a consequence of O binding to the origin of λreplication.

Forty million years of magmatic evolution in the Mariana arc : the tephra glass record

이종만 Univ. of Texas at Dallas 1995 해외박사

마리아나 열도 주변의 10개의 Deep Sea Drilling Project(DSDP) 시추시료에 서 추출한 tephra 유리질은 이 지역에서 4천만년 동안 있어온 폭발적인 화산 활동의 기록을 훌륭하게 제공해 준다. 현무암에서 유문암질에 해당하는 약 1800개 tephra 유리질 입자들의 주성분원소 분석결과를 보고하고 이를 이용하 여 화산활동의 성격과 역사를 고찰하였다. 본 지역에서 특히 활발했던 화산 활동과 관련하여 화산폭발의 정도가 최고에 이르렀던 때가 3회 있었음이 확인 되었다. 폭발적인 화산활동의 가장 큰 피크는 약 18~11 Ma에, 이보다 정도가 낮은 또 다른 2회의 피크가 각각 35~24 Ma 및 6~0 Ma에 있었다. 이들 중 후 기의 두 피크는 다른 서부 태평양 열도들에서 관찰된 화산폭발 활동의 그것들 과 시기가 일치한다. 마리아나 열도의 화산활동은 시작 초기부터 쏘레아이트 질 성분이 아주 우세하였다. 그러나 보니나이트질 성분은 발견되지 아니한다 . 이들 tephra 유리질은, 후기 Miocene (11~7Ma) 동안에 있었던 중에서 high -K 계열의 화산활동을 제외하고는, low~medium-K 계열에 속한다. Tephra 유 리질은 대체로 용암과 성분이 유사하나, 전자는 데이사이트질에서 유문암질 화산 분출물이 완전히 결여된 후자보다 훨씬 넓은 성분 분포를 보여준다. Tephra 유리질은 silica 함량에 따른 bimode 분포를 보여주며, silica 함량이 65~66%인 근처에서 현저한 최저치 혹은 "Daly gap"을 갖는다. 또한 이들은 상당한 결정분화 작용을 받아서 가장 분화가 덜된 유리질은 Fe 값이 높고, Mg # (100Mg/Mg+Fe2+)는 55이다. 현무암질 tephra 유리질은 Al₂O₃ 함량이 16% 이하이고(평균 14. 3%), high-alumina 현무암질이 아니다. 이 사실은 현생 마 리아나 열도의 용암이 15~21%(평균 17. 4%)의 Al₂O₃를 함유한다는 관찰과는 대조를 이룬다. 마리아나 열도의 high-alumina 현무암질 용암은 마그마 용액 성분이 아니라, 아마도 사장석의 집적에 의해서 형성된 혼합체로 해석된다. 전체 tephra 유리질은 감람석-사방휘석-석영 상평형 투영도에서 저압의 반응 선 및 동시 결정선 상에 점시되며, 염기성과 산성 시료는 각각 서로 다른 분 화 경로를 따른다. 염기성 시료의 분화경로는 맨틀기원의 현무암질 마그마의 결정 분화를 반영하나, 산성의 것은 염기성 마그마의 결정분화 혹은 마리아나 열도 지각의 용융에 기인할 수 있다. Tephra glasses retrieved from 10 Deep Sea Drilling Project (DSDP) cores around the Mariana arc system make up a remarkable record of explosive volcanism in the Marianas over the past 40 million years. Major element compositions for approximately 1800 tephra glasses, from basalt through rhyolite, are reported and used to examine the nature and history of this activity. Three maxima of volcanic explosivity, presumably related to especially vigorous volcanism in the Mariana arc, are identified, with the biggest maximum at around 18-11 Ma and two other maxima at 35-24 and 6-O Ma; the two younger maxima are contemporaneous with peaks in explosive volcanism observed for other western Pacific arcs. Explosive arc volcanism has been predominantly tholeiitic since shortly after arc inception; no boninitic glasses were found. The tephra glasses belong to the low- to medium-K suite, except during an enigmatic phase of medium-to high-K explosive volcanism during the late Miocene (11-7 Ma). Even though Mariana tephra glasses are largely similar in composition to Mariana arc lavas, the tephra show a much larger compositional range than that found for subaerial Mariana arc lavas, which apparently lack dacitic to rhyolitic volcanic products. The tephra glasses define a bimodal population in terms of silica content, with a pronounced minimum, or "Daly gap," around 65-66% Si02. Mariana tephra glasses are fractionated, with the least evolved glass being Fe-rich and having a magnesium number (Mg# = lOOMg/Mg+Fe2+) of 55. Basaltic tephra glasses contain <16% A1203 (average 14.3% Al2O3 and are not high-alumina basalts; this contrasts with the observation that modern Mariana arc lavas contain 15-21% (average 17.4%) A1203. The high-alumina basalt lavas of the Mariana arc probably reflect plagioclase accumulation and not liquid compositions. All tephra glasses plot near low-pressure cotectics and reaction curves on the subprojection olivine-clinopyroxene-quartz; mafic and felsic samples define distinct trends. The mafic trend reflects fractional crystallization of mantle-derived basaltic magma, whereas the felsic trend may be due either to anatexis of Mariana arc crust or to fractionation of mafic melts. The tephra glass data reinforce the model that the magmatic evolution of the Mariana arc has been dominated by low-pressure fractionation, perhaps accompanied by anatexis. Episodic changes in melting regime to generate Miocene potassic tephra may be related to changing mantle sources and processes related to episodes of back arc basin spreading. These episodic changes are superimposed on a long-term increase in potassium that reflects progressive metasomatism of the mantle source. Long-term increases in K20 contents for Mariana arc magmas inferred from the tephra glass record are 0.004 wt % m.y.-1 (mafic), 0.011 wt % m.y. -1 (intermediate), and 0.023 wt % m.y.-1 (felsic).

Strategic Bidding Behavior and Cyclival Propertres of Price-cost Margins

이인권 Univ. of Texas at Austin 1996 해외박사

본 학위논문은 두 부분으로 구성되어 있다. 첫 번째 연구는 1986년 Gang of Four(4인방)로 불리는 Kreps, Milgrom, Roberts, Wilson의 논문에서 제시된 비협조적인 게임환경에서 프레이어들의 합리성과 게임의 반복성이 협조적인 결과를 도출할 수 있다는 Folk Theorem의 존재가능성을 통계적으로 검증하고 있다. 4인방의 논문은 여러 차례 반복되는 죄수딜레마타입의 유한반복게임의 초기 단계에서 게임환경에 대한 프레이어들간 정보의 비대칭성(information asymmetry)이 유한반복게임의 균형을 무한반복게임의 균형에 수렴시킬 수 있음을 보여주고 있다. 미국 텍사스 학교우유급식시장의 입찰경쟁 자료의 정밀한 통계분석 결과는 정보의 불확실성에 기초한 tit-for-tat전략이 비협조적인 게임환경의 유한반복게임에서도 경쟁수준이상의 협조적인 결과를 도출할 수 있다는 추론의 타당성을 입증해주고 있다. 경쟁자가 낮은 입찰가격으로 자기가 매년 되풀이하여 공급관을 따낸 지역, 즉 연고지역(incumbent district)을 가로챈 경우, 보복적인 차원에서 경쟁자의 연고지역을 공격적인 입찰가로 빼앗아 가는 tit-for-tat전략을 엄격히 실행해 왔다. 본 연구는 이런 엄격한 tit-for-tat전략이 경쟁업체들로 하여금 연고지역을 존중해주는 협조적인 입찰행동을 유도하고 상대적으로 경쟁업체간의 시장점유율을 안정적으로 유지하며, 경쟁적인 낙찰가 보다 휠씬 높은 낙찰가를 형성하여 경쟁수준 이상의 이윤을 누리게 했음을 밝히고 있다. 두 번째 연구에서는 1963-1987년의 샘플기간 중 미국 180개 제조업체의 패널자료를 활용하여 1986년 Rotemberg와 Saloner의 과점산업 슈퍼게임(oligopoly supergame)의 이론적인 추론인 이윤율(price-cost amrgin)의 역경기순환적인 (counter-cyclical) 특성을 검증하고 있다. 주 아이디어는 과점산업의 경쟁기업들은 경기호황기에 암묵적 담합으로부터 이탈하려는 유인이 높아지는 경우 이를 견제하기 위해서 스스로 간격을 낮추어 경쟁기업들이 담합으로부터 이탈하려는 인센티브를 억제하려 한다는 내용이다. 연구결과는 이윤율의 역경기순환적인 특성을 통계적으로 확인해주고 있다. 두 번째는 경기호황기보다 경기불황기에 이윤율의 역경기순환적인 특성이 더욱 강하게 나타난다는 추론을 검증하였다. 주 아이디어는 경기호황기에 담합으로부터 이탈할 경우 얻는 이익은 상대적으로 가장 크지만 경기불황기에는 담합으로부터 이탈할 경우의 기대손실이 상대적으로 낮기 때문에 담합을 유지하기가 더어렵다는 것이다. 그래서 경기불황기에 이윤율의 역경기순환적인 특성이 더욱 강하게 나타난다는 것이다. 이 추론도 통계적 검증을 통해 확인되었다. 산업별 연구를 한 결과 93개 산업에서 이윤율의 역경기순환적인 특성이 그리고 87개 산업에서 이윤율의 호경기순환적인(pro-cyclical) 특성이 나타났다. 역경기순환적인 특성을 나타내는 산업은 상대적으로 성장이 빠르고 고정비용(fixed cost)의 부담이 적은 산업으로 나타났다. This dissertation investigates the role of the pioneer Korean christians in beginning indigenous Presbyterian churches in Korea and in translating the Gospels into the Korean vernacular. Protestant christianity was introduced in Korea at a time when traditional religions were in a state of decay. The peoples receptivity to Protestant Christianity was closely related to the religious climate, since they were looking for something new that would quench their religious thirst. They were quick to appreciate the gospel. In addition, the instability of the socio-political situation that wae created through a series of criese in the late nineteenth century encouraged people in Korea to turn to Protestant Christianity. Pioneer Korean Christians who came mainly from northwestern Korea established the first indigenous Presbyterian churches, beginning the first Protestant missions in the forbidden land in which bo missionary was allowed to work, Most of these indigenous churches were self-supporting and self-propagating from the beginning. The indigenous Christians also played a renarkable role in translating the Fospels into the Korean vernacular. They produced two indigenous versions of the Gospels, the Yesoo Shung-Kyo version in Manchuria and the Yi Soo-Chung version in Japan, before growth of Protestant Christianity in Korea, laying the biblical foundation for the Korean Protestant church. The foreign missionaries in Korea carried on their work in the indigenous Protestant churches. They did not claim to be the founders of the Protestan church in Korea, bur they played a remarkable role in training indigenous leadership the developing the soul-winning churches.

Hybrid approaches to solve dynamic fleet management problems

김영진 Univ. of Texas at Austin 2003 해외박사

The growing demand for customer-responsive, made-to-order manufacturing is stimulating the need for improved dynamic decision-making processes in commercial fleet operations. Moreover, the rapid growth of electronic commerce through the Internet is also requiring advanced and precise real-time operation of vehicle fleets. Accompanying these demand side developments/pressures, the growing availability of technologies such as AVL (Automatic Vehicle Location) systems and continuous two-way communication devices is driving developments on the supply side. These technologies enable the dispatcher to identify the current location of trucks and to communicate with drivers in real time affording the carrier fleet dispatcher the opportunity to dynamically respond to changes in demand, driver and vehicle availability, as well as traffic network conditions. This research investigates key aspects of real-time dynamic routing and scheduling problems in fleet operation particularly in a truckload pickup-and-delivery problem under various settings, in which information of stochastic demands is revealed on a continuous basis, i.e., as the scheduled routes are executed. The most promising solution strategies for dealing with this real-time problem are analyzed and integrated. Furthermore, this research develops, analyzes, and implements hybrid algorithms for solving them, which combine fast local heuristic approach with an optimization-based approach. Simulation experiments are developed and conducted to evaluate the performance of these algorithms.

A METHODOLOGY TO DESIGN AN AGILE SUPPLY CHAIN BUSINESS MODEL USING OBJECT-ORIENTED APPROACH AND ANALYTIC HIERARCHY PROCESS

김진호 Univ. of Texas at Arlington 2004 해외박사

Supervising Professor: Katherine J. Rogers With the market environment changing dynamically and rapidly, the agility of the business model becomes a survival factor not just a competitive advantage for the success of an enterprise. Agility of a business model means the ability of frequent remodeling with reduced risks and setup-time. In addition, the need of an agile supply chain has been dramatically increased for all kinds of enterprises due to dynamic and unstable market environments. Traditionally, structured or procedure-oriented (PO) modeling methodologies and mathematical optimization modeling approaches have been widely used to design a supply chain, but they have potential limitations to satisfy dynamic demands due to the static properties inherently they have. From this point of view, an object-oriented modeling approach could be a better way to meet these requirements. In this dissertation, a heuristic methodology to design and manage an agile supply chain business model is proposed using object-oriented (OO) modeling approach and analytic hierarchy process (AHP). For the OO designing, UML (Unified Modeling Language), the standardized OO notation by Object Management Group (OMG), is adopted to illustrate the methodology and AHP is utilized for managing the model to efficiently respond to the external changes. The cooperation of OO modeling approach and AHP technique gives the model a synergetic effect for agility. Second, a new application of Group Technology (GT) to the OO model beyond the traditional shop floor level machine-part clustering is introduced. This work is to overcome the potential weakness of traditional methods for identifying classes in the OO modeling field. Third, the knowledge of business rule is included in the methodology to make the model more agile and robust. This work is triggered by the need for work on the integration of business rules with object-oriented paradigm. For this purpose, a procedure for identifying and expressing business rules in the OO model is proposed. Fourth, this dissertation suggests an application outline for adapting this methodology to the entire supply chain based on ‘Supply Chain Operations Reference (SCOR)’ model. Finally, the empirical comparing the agility of OO approach to traditional (i.e. structured or procedure-oriented) approach under the implementing and maintaining the model stage is presented. Some statistical tests and software engineering measures are utilized to validate this comparison results. The results of this study show that OO approach to implement and maintain the supply chain business model has advantages over the traditional approach.

The Effects of Adsorptive Sinks on Human Exposure to VOCs in Indoor Air

원도윤 Univ. of Texas at Austin 1999 해외박사

Indoor surface materials can affect the fate of volatile organic compounds (VOCs) in indoor air. Sink effects associated with sorptive materials may lead to reductions in peak concentrations of VOCs, and hence reductions in occupant exposures. Indoor materials may also serve as reservoirs of VOCs, thus prolonging chemical retention times in indoor environments. However, previous research has been limited to a few chemicals and materials. As a result, attempts to extrapolate adsorptive sink effects to a wide range of chemicals are difficult. This study involved the characterization of interactions between indoor materials and VOCs. This interaction was described using a linear adsorption/desorption model. Adsorption and desorption coefficients were estimated based on a series of 302 experiments involving eight chemicals and twelve materials. Relative humidity, air exchange rate, and inlet gas concentration were varied to evaluate their influence on sorptive interactions. Carpet with an underlying polyurethane pad and virgin gypsum board exhibited the greatest sorption capacity for non-polar and polar compounds, respectively. Most of the sorbed mass for carpet systems was accounted for by the underlying pad and structural backing rather than fibers. Good correlations between sorption coefficients and chemical vapor pressures were observed for most materials. Relative humidity had a significant effect on the degrees of 10 sorption to virgin gypsum board and for a highly soluble chemical (isopropanol) in contact with a carpet system. Air exchange rate and inlet gas concentration has little influence on sorptive interactions. Sorption coefficients from small-scale experiments were evaluated based on experiments completed in a large chamber. Three material combinations and VOCs were used for scale-up experiments. Concentration measurements were predicted reasonably well using sorption coefficients estimated from small-scale experiments. Finally, sorption coefficients were used in conjunction with an indoor air model to determine the effects of adsorptive sinks on occupant exposure to VOCs in indoor environments. The reductions in instantaneous concentrations by adsorptive sinks were significant in several hypothetical scenarios. However, for most cases involving VOCs with high vapor pressures, the influence of sorptive interactions on occupant exposure was observed to be small.

An HPQ-adaptive finite element method for hierarchical modeling of elastic structures

조진래 Univ. of Texas at Austin 1995 해외박사

물리적 현상을 위한 모든 컴퓨터 시뮬레이션(computer simulation)에 있어 서, 첫번째 고려사항은 그 현상을 적절히 묘사할 수 있는 수학적 모델의 선정 이다. 일반적으로, 이러한 모델선택은 해석자의 주관으로서, 당면한 문제에 대한 해석자의 경험, 판단 및 문제와의 친숙도에 기초한다. 수치적 해석의 성패여부가 선정된 모델 그 자체의 적합성에 절대적으로 의존하는 것은 명확 한 사실이다. 그럼에도 불구하고, 전산과학(computational science)의 역사 를 살펴볼 때, 모델의 적합성 여부는 시뮬레이션이 완료되어 얻어진 결과가 다른 자료와 비교된 뒤에야 판단되었다. 이 논문은 얇은 탄성체(thin elastic bodies)의 응력이나 변위해석과 같은 중요한 문제들에 대한 적절한 모델의 순응적 선택(adaptively selecting)이란 최신의 개념을 탐구한다. 이러한 박판 구조물의 종래의 모델들은 plate나 sh ell이론에 근거하며, 치명적인 error를 수반한 시뮬레이션을 낳을 수 있다. 본 연구는 변위장(displacement field)의 박판두께 방향으로의 polynomial 의 차수 q를 변화시킴으로써 얻어지는 계층적 모델(hierarchical models)의 개발에 관한 것이다. 박판의 중립면(mid-surface)에서는 기존의 hp-version 유한요소법이 적용된다. 본 연구에서 계층구조의 최상위 모델은 3차원 선형 탄성이론이다. 여러 모델들의 quality를 평가하기 위해서, 후 error 평가(a- posteriori error estimation)의 유도가 필요하고, 이 평가방법은 해석자로 하여금 적절한 모델을 자동적으로 선택케하는 adaptive strategy에 필수적이 다. 더욱이, 최소의 자유도(degree of freedom)로써 구조물을 해석하기 위해 서 구조물의 각기 다른 영역에 각기 다른 모델들이 적용된다. 이 연구는 공학상의 중요한 문제들에 대한 자동적 모델선정(automatic mode l selection)과 그들의 적합성 평가를 위한 후 error 평가를 systematic하게 처리하는 절차를 제시한다. In every computer simulation of a physical system, the first consideration is the selection of an appropriate mathematical model, a collection of mathemat-ical constructions, characterized often by partial differential equations, which portrays, in mathematical terms, theories on the behavior of the system. In general, the choice of the model is the perogative of the analyst and is made on the basis of experience, judgement, and familiarity of the analyst with the problem at hand. Obviously, the success of a numerical simulation of a physi-cal event depends strongly on the suitability of the model itself, notwithstand-ing the quality of the solution obtained from a specific model. Nevertheless, throughout the history of the computational sciences, the appropriateness of various classical models is rarely contested until after a simulation is made and results are compared with other data. This dissertation explores the idea of adaptively selecting appropriate models for an important class of problems; the analysis of stresses and de-formations of thin elastic structures. Classical models of such structures are based on so-called plate and shell theories, and can lead to simulations that are significantly in error. The study considers the development of hierarchical models in which mathematical models of varying sophistication are produced parametrically by changing the assumed variation in the displacement fields in a thickness direction identified with the geometry of thin bodies. This is referred to as the Q-approximation, Q being the order of the polynomial approximations through the thickness. Elsewhere, hpversion finite element approximations are employed. The highest level of sophistication in the hierarchy is defined as the three-dimensional linear elasticity in this study. To assess the quality of various models, methods for a posteriori error estimation are derived. These produce estimates that are used in an adaptive strategy that enables the analyst to automatically select an appropriate model for the problem under study. Moreover, different models may be used at different portions of the structure to capture the behavior with a near minimum number of degrees of freedom. It is found that so-called locking phenomena can occur in such models owing to the disproportionate scale interacting in certain convergence processes. A study of locking and of methods to counter it is also included in this work. The study presents systematic computational procedures, based on mathematical analyses, for the automatic selection of models and the a posteriori evaluation of their suitability for a significant class of engineering problems.