蛋播视频一区,无码鲁丝一区二区,精品 久久 五月天,国产老熟女,五月草草在线观看,中文日韩欧美,情色一区二区三区,欧美日韩亚洲激情在线,亚洲制服在线香蕉

Optimal Update Policy for CRM Databases in the Present of Dependent Errors

講座時間：9月24日（周二）9：30-11：30

講座地點：25樓A區(qū)3層C教室

主講人：姜正瑞

主講人介紹：

姜正瑞，管理科學(xué)博士，南京大學(xué)商學(xué)院營銷與電子商務(wù)系教授，博士生導(dǎo)師。在2019年加入南大之前，任美國愛荷華州立大學(xué)（Iowa State University ）商學(xué)院的信息系統(tǒng)與商業(yè)分析教授和Thome講席教授。在國際頂尖期刊如

講座內(nèi)容:

Data is becoming an increasingly important asset to today’s organizations. Challenges in the age of big data, such as high volume and high velocity, call for better database maintenance strategies. This research focuses on deriving the optimal update policy for CRM databases with customer records. To better support business applications and reduce the costs of stale data, CRM databases need to be updated often. However, frequent updates can drain computing resources and cause interruptions to business operations. In this study, we take a realistic and details view of the type of errors resulting from stale data and the costs of such errors. By trading off the cost of updates and data staleness, we propose methodologies to decide when or how frequently to update a consolidated database. We consider both databases that are still in its growth state and those who have achieved a steady state. We find that a fixed update cycle policy is guaranteed to be optimal only for steady state database under an infinite time horizon. When the time horizon is finite, the optimal policy features either equal-length update cycles or update cycles that differ in length by exactly one time interval. For databases that are in a growth state, the update cycles tend to decrease in length over time. We expect our policy to outperform other policies considered in the prior literature, and lead to more efficient database maintenance and better decision support for organizations.

封面圖制作 / 劉洋

底圖制作 /楊仕程

圖文編輯 / 趙博洋