台灣資訊業早期從國際貿易起家，出口一直是台灣與其它國家往來的主要型態，隨著國際營運經驗的累積和各種因素的改變，公司開始考慮加深涉入國際市場，於是該如何採取下一步的進入策略成為一大議題。過去進入策略的研究雖多，但多偏向橫斷面的研究，研究重點主要在各項因素對進入策略之影響，對公司現階段進入策略如何影響下一階段進入策略的選擇卻無研究，本研究的目的即在彌補此一缺口。 本研究嘗試以動態的觀點來探討進入策略，即研究公司歷經現階段的進入策略到底學到或得到些什麼？而學到的經驗或得到的資源如何幫助公司去做下一階段進入模式的選擇？本研究的主要問題有二：（一）當台灣資訊業在決定國際市場進入策略時，哪些因素會影響此一決定？（二）當海外進入策略有變遷時，上一階段的進入策略學到了什麼事情？獲得什麼資源？這些經驗與資源如何影響下一階段進入策略之選擇？ 本研究的主要命題分為七大類，命題1主要是陳述當公司開始國際化時，有哪些因素會影響公司選擇進入策略，此命題回答了問題1；命題2主要是描述有哪些因素會影響進入策略由出口階段轉為子公司階段；命題3則是出口轉為合資子公司的原因；命題4描述公司為何採取由出口轉為購併的決策；命題5則是設立子公司之後，有哪些因素會影響公司考慮在海外設廠；命題6則是公司設立第一子公司之後，接著設立其它子公司的原因；命題7則是公司先設立工廠之後，接著設立子公司的原因。上述命題2至命題7則可以回答本研究的第二個問題。 本研究的主要結論如下： 一、當母國市場銷售潛力低、海外市場銷售潛力高時、公司缺乏資金、人才、客戶、國際營運經驗時、母國生產成本低廉時，資訊業的公司傾向採用出口； 二、出口轉為銷售子公司或合資子公司的考量因素有拓展業務、減少通路階層、提高服務品質、具有國際營運的人才、存在長久合作的客戶等； 三、出口一段時間之後，公司直接在當地購併工廠的原因是想要快速獲得技術、品牌、與通路； 四、子公司轉為設廠的考量因素有母國的生產成本、地主國的關稅、當地組裝就近服務客戶等因素； 五、在文化相近的市場裡，第一子公司的設立經驗可以對往後設立其他子公司有幫助； 六、公司在地主國以工廠的型態持續一段時間之後，會因為地主國的市場漸漸蓬勃而設立銷售子公司。

近年來，由於Black-Scholes選擇權評價模式的發展，利用套利定價理論來研究利率期限結構和利率性或有請求權定價的文獻有增多的趨勢，而學者們建構利率期限結構模型的方法是由瞬間即時利率遵守某些隨機過程開始，然後推導出在風險中立的情況下，此一波動過程所演變而成的債券價格。例如Cox,Ingersoll and Ross(l985)發表的一般均衡模型，以及Longstaff and Schwartz(1992)根據Cox,Ingersoll and Ross(1985a,b)的單因子模型，再加上即期利率的變異性，而導出一雙因子模型。 在投資人決定購買何種到期期間的證券以及公司決定用長期債券或短期債券融資之前，都應先瞭解利率期限結構，並預測未來利率走勢。本研究擬針對一些著名的單因子動態利率模型以及Longstaff-Schwartz雙因子模型加以分析，並驗證在我國應用的可行性。 研究結果發現，單因子模型中以CIR SR及B-S模型的配適度較佳，顯示台灣利率走勢長期將會回歸到一平均利率，亦即國內的貨幣市場價格風險將得以降低。 近年來有許多學者另行引進許多變數於模型當中，以增加捕捉動態利率的能力，本研究藉著L-S模式，引入另一變數－短期利率的變異性，估計結果可知此變數亦會影響國內貨幣市場之利率結構，並證明了利率之變異性亦為或有請求權定價的重要考慮因素。

由於企業環境劇烈的競爭及快速變化，工作方式將以團隊合作結合群體的智慧取代以個人力量獨力完成，而群組軟體透過電腦軟體及網路能促進組織成員之溝通、協調及合作。 　　本文提出動態團隊產生引擎的觀念，係以解決問題及顧客滿意為導向，根據問題的難易程度找出解決問題所需專長，形成團隊在網路上同時相互討論共同解決問題，並依此觀念應用群組軟體建立雛型。利用這種方式，組織內專家可以突破空間之限制，在網路上解決問題，不需來回奔波，可增加效率及生產力。除此之外，可迅速產生適當人員解決問題可避免大才小用、縮短反應時間、提升組織的彈性及應變能力等。本研究並針對如此的應用，對組織可能產生衝擊作一初步探討。 / Due to the hard competition and dramatic changes in business environment, Jobs and problems will be solved by team work instead of single person. Using groupware on computer network can improve communication, coordination and cooperation among team members. 　　In this research, a dynamic team generation engine is proposed to organize a team based on each problem. The members for each team depend on the problem's level and domain knowledge it needs in the first time, and will be changed when necessary. The team members will discuss the problems through computer network until the solution is found. Based on this kind of concept, a prototype is built, and a small ambulance case will be studied and showed how to solve the problems by this way. 　　Taking the advantage of groupware, experts within the organization can solve the problems more efficiently without being restricted by time and space, which raises the productivity of organization. Moreover, dynamic team generation engine can help to search for adequate experts inside the organization to solve problems, reduce the response time raise the flexibility of the organization and the capability to adjust to changing circumstances. Influences on the management side by the application of groupware to the organization will also be discussed in this research.

本研究討論以負債導向之退休基金的資產配置模型，並以股票型風險性資產為主要配置標的。隨機控制模型在推導過程中相當繁瑣，經常得不到封閉解，本研究之優點為，實際導出多項資產標的下之一般化封閉解，可進行財務經濟推論，直接得到不同參數對基金之影響，輔以台灣公務人員退撫基金第4次精算報告，為實證研究對象，接著加入投資限制之情境分析與模擬，得到結論。最適提撥隨著正常成本及給付上升而提高，若退休基金於當期有殘餘基金，則可因由投資獲利而少提撥部分資金。回饋函數之最適解同時權衡反應未來之精算正常成本與當期給付，正常成本上升而增加風險性投資，而因當期給付上升而減少風險投資趨於保守。若股票市場報酬率大於利率，投資者將增加股票之比例，以增加投資效果，投資者將對市場股票型資產同時做多空操作，進行避險。反之，隨著短期利率上升後，投資於股票部位將會漸漸移入現金持有，減少股票型風險性資產佔總資產之比例。

隨著總體環境不確定性日益增加，企業是否具有因應變動的能力，做出符合環境變動需要之回應，進而在高度競爭的環境下維持競爭優勢，已顯得格外重要。2008年發生的金融危機，對許多產業的廠商造成極嚴峻挑戰。本研究將針對高科技的DRAM產業進行研究，該產業具有高資本密集、高技術密集以及景氣循環明顯等特色。因此，具備因應外在環境變動的能力，對於廠商能否永續經營，扮演了非常關鍵的角色。 故本研究以DRAM產業中具代表性企業的採購組織為例，透過相關文獻探討，以動態能耐的「組織與管理程序」、「地位」及「路徑」三大構面，對個案公司進行動態能耐的辨識，並透過採購策略的擬定與執行，進一步對採購績效產生影響作一深入研究。然而，一個具有動態能耐的採購組織，透過採購策略達成採購績效的過程中，將形成該採購組織的獨特經驗，不論此經驗是好是壞，皆會促使該組織進一步的改善與學習，採購組織的動態能力亦將隨之改變，並再次調整其採購策略。本研究特別強調在金融危機前後，採購組織其動態能耐的調整與改變，以因應外在環境的劇烈變動。 本研究採用個案研究與專家深入訪談，並以次級資料加以佐證。本研究所獲得的研究結論如下： 結論一：外在環境、時間發展、組織文化、組織策略改變以及高階領導人態度，是採購組織發展動態能耐的關鍵因素 結論二：非財務面績效指標的量化，可以讓組織進行更公平的績效考核 結論三：組織具備良好的動態能耐是制定與執行正確採購策略的關鍵因素 結論四：正確的採購策略擬定與制定有助於採購績效的提升 結論五：企業所擁有的獨特動態能耐愈多，愈有助於企業快速調整其採購策略，以因 應快速變化的外在環境，維持企業之競爭優勢。 最後本研究進一步提出後續在學術研究領域以及實務工作相關之採購策略發展建議。

個案公司自1985年創立迄今已超過25年，主要從事個人保護安全工作帶配件的金屬加工製造公司。在歷經貿易商、虛擬工廠、研發製造、品牌行銷等階段的發展後，於2008年遭逢金融海嘯的衝擊。此時公司開始思考如何進行策略重定位，並透過一連串營運活動的轉變，從過去的利基市場切入到以主流市場為主的市場定位，強調以物美價廉、快進快出為公司主要的營運策略，期望藉此挽回快速流失的市占率。海嘯後歷經兩年的策略調整，個案公司的策略也隨著外部需求市場的質變，開始重新調整公司整體的策略定位(從利基走向主流)，並根據策略重定位回頭調整內部組織結構的運作，以利策略的執行。總結上述一連串的轉變，背後支持個案公司得以成功轉型的關鍵在於公司動態回應市場變化的能力，進而形成維持持續成長的動態能力。準此，本研究主要目的即在探討個案公司如何有效運用環境、策略與結構這三層構面彼此的最適互動，來協助其達到成長的目的。 / The case of corporate in this paper is built for 25 years since 1985, it designed and manufactured a range of lifting accessories. The company went through trader, virtual factory, R&D and brand marketing, but happened the financial crisis in 2008. As such, it decided to reconsider the strategic position in the industry. In order to raise the losing market share, the strategic position is focused on mainstream market from niche market through a series changing activities, and the main business strategy pay attention on price, quality and turnover speed. After the financial crisis, the company adjustment its organizational structure depend on strategic position. Furthermore, the key success factor of transformation for the company is the dynamic adaptive capability, and the long term growth path also depend on it. Consequently, the topic of this paper discusses the fit among environment, strategy and structure, which enables the company remains continued competitive advantage.

在這篇論文裡，我們要討論的是在多維度的時間刻度下橢圓型動態算子和拋物型動態算子的極大值定理，並藉此得到一些應用。事實上，我們是將微分方程及差分方程裡的極大值定理推廣至所謂的動態方程中。 / In this thesis, we establish the maximum principles for the elliptic dynamic operators and parabolic dynamic operators on multi-dimensional time scales, and apply it to obtain some applications. Indeed, we extend the maximum principles on differential equations and difference equations to the so-called dynamic equations.

本研究的目的是探究動態能力的本質，並進一步了解動態能力的發展過程。為了解企業如何其發展動態能力，本研究採用個案分析，發現企業發展動態能力須經歷的四個階段，並提出一整合模型描述動態能力發展之過程:企業可經由學習機制發展其動態能力，包含知識獲取以及知識轉化兩個階段;若環境動態程度較高，將促進企業獲取知識，從而加快學習機制。此外，社會資本和管理能力的提升也會影響動態能力的發展。 / The purpose of this study is to understand the nature of dynamic capabilities, and to have a clearer picture of the process of developing dynamic capabilities. A case study methodology is conducted to illustrate the pattern of how firms develop dynamic capabilities. After collecting and analyzing the data, four propositions and an integrated model are proposed. It is found that firms develop dynamic capabilities through the two-stage learning mechanism, knowledge acquisition and knowledge conversion; high level of environmental dynamics perceived by firms would motivate firms to acquire knowledge, which accelerate learning mechanism. Moreover, social capital and managerial capability upgrading play roles of moderator during the development of dynamic capabilities. Four contributions are offered in this study. First, this research highlights the ambiguity on definition and clarifies it after an overall discussion. Second, this paper extends and integrates findings from past research, which help to enhance the understanding of the mechanism of developing dynamic capability. the third contribution of this study is the combination of learning mechanism and two moderators, social capital and managerial capability upgrading. Fourth, while cases from advanced countries have been widely discussed, this study focus on companies from newly industrialized countries, which enables this study to collect different managerial experience and extend more insights. There are several research limitations in conducting this study. Firstly, the use of qualitative research limits the research result in the specific industry chosen in this study; some propositions may be universal, but not all of them can be applied into each industry. Secondly, the timeline covered in this study is not long enough since the development of dynamic capabilities requires long time efforts.

廠商間的行動與回應是競爭的基石 (Chen and MacMillan, 1992; Chen and Miller, 1994; Smith, Grimm, Chen, and Gannon, 1989)。行動是廠商分析競爭環境後，所得出競爭優勢的一種展現；行動也是分析競爭者後，針對所預期競爭對手反應，在擬定之替代性方案中所做的選擇的實踐。 動態競爭理論對競爭互動的討論已略有共識 (Chen and MacMillan, 1992; Chen, 1996)，但其研究標的多為市場中競爭者之間的攻擊與反擊，對於以合作模式所萌發的廠商間關係，究竟是如何在合作關係之上發展出競爭，學術的研究仍在逐步累積之中 (Chen, 2002; 2008; Brandenburger and Nalebuff, 1996; Peng and Bourne, 2009; Peng, Pike, Yang, and Roos, 2012)。 本論文主張，代工廠自創品牌時與品牌商的競爭互動即是一套套合作與競爭的策略組合，廠商的決策與行動反映了廠商所偵測到的環境變化與所自我評估的能耐水平，並納入競合思維，而推測彼此之競合互動的模式時則可以賽局理論為依據(Camerer, 1991; Brandenburger and Nalebuff, 1995; 1996; Day and Reibstein, 1997; Chen, 2008)。 本研究探討之研究問題包括: (一) 究竟代工廠在什麼情況下會考慮進行垂直整合來自創品牌？(二) 代工廠決定自創品牌後，其自創品牌的競合策略行動與考量為何? 本研究的研究設計採Eisenhardt (1989) 之歸納性個案研究法，選取的個案公司為10家名列台灣20大國際品牌之廠商，資料蒐集來自多重來源之初級與次級資料，包含與企業主與高階管理者之面對面訪談，以及網站資訊、報章報導、公司簡介與年報財報等等。本研究的分析單位是廠商自創品牌的決策及自創品牌的行動與回應。 針對研究問題一，本論文發現對自創品牌之代工廠來說，產業環境中的變化會使得代工廠改變其目標內涵與目標水準。產業環境可歸納出兩個子構面，一個是自創品牌的壓力，另一個是自創品牌的誘因。誘因與壓力的交互影響決定了代工廠自創品牌的意願；其中自創品牌的誘因對自創品牌的意願影響程度較高。 針對研究問題二，本論文發現代工廠自創品牌行動的作法是根據大環境，代工廠與品牌商的矛盾，代工廠自身的能力、價值觀，以及品牌商的價值觀，以築高品牌商回應障礙為目標去反向推導出適當的自創品牌行動。在不同的產業環境與廠商關係之下，代工廠商會以不同行動與路徑來組合動態競合策略，以求取組織間競爭與合作的平衡。 / Action and response is the capstone of inter-organizational competition (Chen and MacMillan, 1992; Chen and Miller, 1994; Smith, Grimm, Chen, and Gannon, 1989)。Action manifests a firm’s competitive advantage that is derived from in-depth scanning and analysis of the environment; action also realizes the decision that derived among strategic alternatives and analysis of competitors ‘response. In the last few decades, competitive dynamic researchers have made some insightful discussion about action and response in inter-organization competition (Chen and MacMillan, 1992; Chen, 1996). However, the majority of the aforementioned researches focus on action and response among competitors; the study of inter-organization relationship among vertical value chain, or any other cooperative governance mode, is still mildly increasing and called for (Chen, 2002; 2008; Brandenburger and Nalebuff, 1996; Peng and Bourne, 2009; Peng, Pike, Yang, and Roos, 2012). This study submits that the competitive interactions between contract manufacturer and brand owner can be depicted as an array of combinations of competition and cooperation. The action and response that a firm chooses to take reflect firm’s thought on the change in the environment and the competency in the firm. Basing on game theory, this study incorporates the concept of competition-cooperative to infer and identify the possible interaction modes in a vertical relationship, particularly in the setting of cross-border contract manufacturing (Camerer, 1991; Brandenburger and Nalebuff, 1995; 1996; Day and Reibstein, 1997; Chen, 2008). This study aims to extend the discussion on competition and cooperation in a vertical relationship. In essence, two research questions are investigated. First, responding to the pressure and opportunity in the operating environment, what are the environmental and capability conditions that motivate a contract manufacturer to consider building own brand? Second, what are the own-brand building contract manufacturer’s concerns and modes underlying each competitive-cooperative action and reaction? In order to explore the nature of the strategic thinking behind contract manufacturer’s decision, this study employs a multiple-case design (Eisenhardt, 1989). Ten contract manufacturing firms, shortlisted from the list of Taiwan top 20 international enterprises, are selected for field studies. Data are collected from multiple sources, ranging from primary data and secondary data, via conducting interview with executive and reviewing through company websites, newspaper, company document, and published financial reports. In this thesis, the focus is the strategic thinking underlying the action and response of contract manufacturer who decides to go build own brand. This study discovers that the change in the environment leads the firm to alter the corporate objective and level of it. There are at least two analytic dimensions behind the strategic decision for a contract manufacturer to build own brand, one is the cost pressure existing in the inter-firm transaction; the other is the opportunity in the market. The interaction between transaction cost pressure and market opportunity impacts contract manufacturers’ level of willingness to build own brand. In this study, we find that market opportunity demonstrates a greater impact on contract manufacturer’s willingness to build own brand. Furthermore, this study discovers that contract manufacturers pave their way of brand building based on the analysis of action and response. An action is determined by identifying change in the environment, anticipating level of conflict in the vertical relationship, and gauging the level of competency and value held by transaction partners in a vertical relationship. Additionally, this study discovers that contract manufacturer formulates a dynamic co-optition strategy that employ an array of action and response as well as a variety of paths on the way to build own brand, with the intension to further strike a good balance between the inter-organizational competition and cooperation.

MapReduce 是目前最熱門的雲端技術之一，用來處理大量資料，不論資料探勘、非結構化的紀錄檔、網頁索引處理及其他需要大量資料處理的科學研究，都可透過 MapReduce 得到極佳的執行效率。MapReduce 為一分散式批次資料處理程式框架，將一個工作分解為許多較小的 map 任務以及 reduce 任務，由map 處理每個小問題，再由reduce將問題彙整，得到最終的結果。 Hadoop 是一個開放原始碼的 MapReduce 架構，並且被廣泛地應用在以大規模資料運算為主的雲端計算。Hadoop有一個非常重要的元件稱為scheduler ，是 hadoop的中樞，負責調度、指派任務和資源分配的優先順序。Scheduler的任務選擇與分配方式，將會影響 MapReduce 工作的執行效率與整個叢集的使用率，目前Hadoop預設的scheduler是將任務以先進先出(FIFO)的方式進行排程。提升MapReduce運算效能的挑戰之一為如何適當的分配Mapper 和 Reducer給雲端裡的每個節點來執行。儘管過去已經有許多改善MapReduce運算效能的研究，但是大部分的方法在實際的運作中，仍存在很多的問題，如工作節點的動態負載、data locality的問題，計算節點的異質性等等。我們發現目前Hadoop對於這些問題並沒有妥善處理，並且在相關的情況下，整體效能仍有改進空間。 我們提出Data Locality Driven Scheduler(DLDS)的方法，並實踐在 Hadoop上，試圖提高scheduler的效能。我們設計不同的實驗，比較DLDS在不同狀況下和其他的排程演算法的差異。實驗結果顯示，透過提高資料的地域性，平均可提昇10% 至 15% 的效能。 / MapReduce is programming model for processing large data set. It is typically used to do distributed computing on clusters of computers such as Cloud computing platform. Examples of bit data set include unstructured logs, web indexing, scientific data, surveillance data, etc. MapReduce is a distributed processing program framework, a computing job is broken down into many smaller Map tasks and a Reduce task.Each Map task processes a partition of the given data set and Reduce aggregates the results of Maps to produce final result. Hadoop is an open-source MapReduce architecture, and is widely used in many cloud-based services.To best utilize computing resource in a cloud server, a task scheduler is essential to assign tasks to appropriate processors as well as to prioritize resource allocation. The default scheduler of Hadoop is first-in-first-out (FIFO) scheduler which is simple but has a performance inefficiency yet to be improved. Although there have been many researches aiming to improve the performance of MapReduce platform in the past year, there still have many issues hindering the performance improvement, such as dynamic load balance, data locality, and heterogeneity of computing nodes. To improve data locality, we propose a new scheduler called Data Locality Driven Scheduler (DLDS) based on Hadoop platform. DLDS improve Hadoop's performamce by allocating Map tasks as close as possible to the data block they are to process. We evaluated the proposed DLDS against several other schedulers by simulation on an 8 nodes real Hadoop system. Experimental results show that DLDS can improve data locality by 10-15%, which results in a significant performamce improvement.