作者: 許 欣怡

  • 110.3.29.演講-Before It’s Too Late: Product Recall Delays and Policy Design

    講    題:Before It’s Too Late: Product Recall Delays and Policy Design

    主 講 人: 李曉惠Hsiao-Hui Lee教授

    (國立政治大學資訊管理學系)

    主 持 人: 陳文智  教授

    主辦單位:交通大學工業工程與管理系

    時間:110年03月29日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

    When a safety defect occurs, manufacturers often use product recalls to mitigate potential consequences. Although consumers expect on-time recalls for product defects, anecdotal examples suggest that firms may be passive in investigating potential defects and/or severely delay their recall decisions. In this paper, we incorporate a Bass-diffusion product cycle model into firms’ investigation and recall decisions when defects occur, and provide social planners with various instruments to deter long delayed recalls. Our theoretical model reveals three main results. First, not all firms will delay recalls; the firm that suffers more negative impacts from external channels and has a relatively high margin-to-recall-cost ratio will consider a delayed recall. Second, a firm that will consider a delayed recall also exerts less effort in investigating product defect. Third, an earlier defect notice time may not necessarily lead to an earlier recall time and/or higher investigation effort; these decisions depend on a firm’s learning on product defect and penalty for delay. Our model not only helps us understand how firms make their recall timing decisions, but also offers governments and regulation bodies new instruments (e.g., investigation efforts, penalty design, information disclosure, firm supervision) to help encourage firms be proactive when a defect occurs, thereby reducing potential casualties associated with delays in a recall process.

    演講性質:學術研究專題

    歡迎聽講

  • 110.3.18.演講-「深度學習,三思而後行」

    講    題:「深度學習,三思而後行」

    (Deep Learning, Think Twice)

    主 講 人: 桑慧敏  教授

    主 持 人: 巫木誠  教授

    主辦單位:交通大學工業工程與管理學系

    時間:110年3月18日(星期四) 13:20~15:00

    地    點:管理二館520室

    演講摘要:

    先以桑慧敏廣播,podcast,「一葉之秋, 一生受用的數據分析」中第17講
     「人工智慧」中的歌曲(旋律:世界真是小小小)開頭:
     「有人真靈巧,機器稱電腦。人腦來做主,電腦做苦勞。
    不偷懶不疲勞,這個小電腦,人工智慧多美妙!
    見賢思齊,舉一反三,三思而後行,妙妙妙。
    明察秋毫,反省改善,人工智慧真奇妙!」
    唱出所謂「人工智慧」的涵義。
    接著以青光眼與先進半導體瑕疵偵測為案例說明深度學習的架構。深度學習中所有關鍵方法(如 Loss function, optimization, performance measure, residual learning, block, epoch, and batch等),能讓聽者輕鬆學會,且終身牢記

    演講性質: 學術研究專題

    歡迎聽講

  • 110.3.22.演講-智慧決策分析與應用

    講    題:智慧決策分析與應用

    主 講 人: 林東盈Dung-Ying Lin教授

    (國立清華大學工業工程與工程管理學系)

    主 持 人: 陳文智  教授

    主辦單位:交通大學工業工程與管理系

    時間:110年03月22日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

    藉由交通運輸路/海/空運優化、生產排程與智慧排班等案例,探討作業研究(Operations Research)之核心理論與實務應用,並分析業界導入作業研究等優化系統時,可能遭遇的問題與潛在解決方案。

    演講性質: 學術研究專題

    歡迎聽講

  • 110.3.15.演講-驚喜和不確定性的效果:以真實市場實驗探討消費者行為

    講    題:驚喜和不確定性的效果:以真實市場實驗探討消費者行為

    主 講 人: 陳瑀屏Yu-Ping Chen 教授

    (國立臺灣大學國際企業學系)

    主 持 人: 陳文智  教授

    主辦單位:交通大學工業工程與管理系

    時間:110年03月15日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

    消費者行為的研究指出,比起預期中的促銷活動,意料之外的驚喜折扣更能夠提高消費者的滿意度,其效果更能夠延長到活動結束之後。另外,比起確定的折扣,消費者更喜歡帶有不確定性的促銷活動,即使兩者的期望值相同,甚至不確定性的促銷活動期望值更低,消費者仍有更高的願付價格及參與率。雖然這些研究建立了消費者行為的心理機制及相關的模型,並對未來的行銷活動有很大的啟發,然而,這些研究多半在實驗室中進行,建立在假設性的研究設計上,並只觀察受試者當下的滿意度及行為。為了補充文獻上的不足,我們分析真實世界中的消費數據,也透過市場實驗的方式,和相關的廠商合作,在真實世界中進行實驗,並持續追蹤消費者後續的回購行為,以探討驚喜和不確定性等促銷方式,如何影響短期和長期的消費者行為。

    演講性質:

    •  學術研究專題

    歡迎聽講

  • 110.3.8.演講-Dynamic Inventory Optimization with Learning and Model Ambiguity

    講    題:Dynamic Inventory Optimization with Learning and Model Ambiguity

    主 講 人: 莊雅棠Chuang,Ya-Tang 教授

    (國立成功大學工業與資訊管理學系)

    主 持 人: 陳文智  教授

    主辦單位:交通大學工業工程與管理系

    時間:110年03月08日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

    This research concerns optimal inventory control in the presence of model ambiguity and statistical learning. Specifically, decision makers, on the one hand, face inventory control problems where parameters of the demand distribution are not known a priori, and need to be learned using right-censored sales data. On the other hand, the decision makers fear that the underlying model may be misspecified and attempt to find a robust policy against model ambiguity. Inventory control problems with learning are usually modeled using Bayesian dynamic programming (BDP). Under the Bayesian paradigm, it is assumed that the family of the demand distribution is known and data is generated from the presumed demand distribution. This assumption is however, not always satisfied for most practical problems. Another modeling approach, robust optimization, is usually used when the model is misspecified, and the decision makers hence solve a worst-case objective to hedge against model ambiguity. In this approach, data is essentially of no value when it comes to learning the model. Motivated by this concern, the goal of this research is to (i) develop new modeling frameworks that allow the decision makers to remain robust with respect to model ambiguity while also learning at the same time, and (ii) understand the effects of learning and robustness on the optimal objective values and decisions of the proposed model. The primary focus of our analysis is on establishing structural results of the decision makers’ optimal decisions. Our main result shows that the optimal decision can be expressed as the sum of a myopic decision plus a (non-negative) learning boost, minus a (non-negative) robust adjustment and an (non-negative) interaction adjustment. Moreover, through this representation, we find that the famous “stock more” result could be sometimes revised to stock less even if the benefit from learning is increasing.

    演講性質: 學術研究專題

    歡迎聽講

  • 109.12.22.演講-半導體設備商的工作點滴

    講    題:半導體設備商的工作點滴

    Work experience sharing in semi equipment company

    主 講 人:郭奇文 先生 (台灣應用材料)

    Chi Wen Kuo

    主 持 人:劉建良  教授

    主辦單位:交通大學工業工程與管理系

    時    間:109年12月22日(星期二) 13:20~15:20  

    地    點:管二館 506 室

    演講摘要:

     半導體設備商的工作點滴

    演講性質:職場經驗分享                歡迎聽講

  • 109.12.21.演講-Clustering for Unsupervised Learning: from Manufacturing to Diverse Data

    講    題: Clustering for Unsupervised Learning: from Manufacturing to Diverse Data

    主 講 人:廖崇碩 教授

          (清華大學工業工程與工程管理學系)

    主 持 人: 洪暉智 教授

    主辦單位:交通大學工業工程與管理系

    時間:109年12月21日(星期一) 13:20~14:20

    地    點:管二館520室

    演講摘要:

    In the new age of big data analytics, clustering has become more and more important for analyzing data, particularly with its application to unsupervised machine learning due to the lack of quality labelled data. Clustering has been widely exploited in a variety of fields, while there have been many different types of clustering algorithms widely-studied in the literature. In this talk, we will first introduce a new dynamic density-based clustering algorithm with its application to time-series data. Next we show numerical experiments over real-world data to demonstrate its effectiveness. 

    演講性質:學術研究專題 歡迎聽講

  • 109.12.14.演講-使用作業研究方法於醫療應用

    講    題: 使用作業研究方法於醫療應用

    (Using Operations Research Approaches on Healthcare Applications )

    主 講 人:陳平舜 教授

    (中原大學工業與系統工程學系)

    主 持 人: 洪暉智 教授

    主辦單位:交通大學工業工程與管理系

    時間:109年12月14日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

     This talk will cover a brief introduction of operations research methodology and three applications on healthcare. The first application applies system simulation to construct patients’ appointment model and implements four appointment scheduling policies—namely, constant arrival, mixed patient arrival, three-section pattern arrival, and irregular arrival—in the case ultrasound department. The second application uses mathematical programming to model a nurse scheduling problem, which contains constraints of government regulations, hospital regulations, and nurse preferences. This application applies two meta-heuristic algorithms, bat algorithm and shuffled frog-leaping algorithm, to search for a near-optimal solution under different testing instances. The third application uses the pooling-resource concept to allocate medical staff among collaborative hospitals and develops three heuristic algorithms integrated with the particle swarm optimization algorithm in order to determine an optimal medical staff’s monthly schedule.

    演講性質:學術研究專題 歡迎聽講

  • 109.12.8.演講-聊天機器人應用

    講    題:聊天機器人應用Chatbot Applications

    主 講 人:李青憲ching-hsien lee

    (工業技術研究院 巨量資訊科技中心)

    主 持 人:劉建良  教授

    主辦單位:交通大學工業工程與管理系

    時    間:109年12月8日(星期二) 13:20~15:20  

    地    點:管二館 506 室

    演講摘要:

     大家好,我是李青憲,任職於工研院巨資中心,專長是文字分析與機器學習

    演講性質:職場經驗分享                歡迎聽講

  • 109.12.7.演講-服務網路與作業研究之研究經驗與心得

    講題: 服務網路與作業研究之研究經驗與心得

    (On the Research Experiences in Service Network and Operations Research)

    主 講 人:王逸琳 教授

    (成功大學工業與資訊管理學系)

    主 持 人: 洪暉智 教授

    主辦單位:交通大學工業工程與管理系

    時間:109年12月07日(星期一) 13:20~15:10

    地    點:管二館520室

    演講摘要:

      本演講專注於利用作業研究概念與基礎知識,包含線性規劃、敏感度分析、指派問題與網路規劃等。並說明如何利用作業研究建構實際問題與尋求最佳解。網路服務中各種做法提出部分觀點與心得。並由研究中提出最佳結論。  

    演講性質:學術研究專題 歡迎聽講