標題: IT服務管理:運用情境認知之知識支援於事件管理
IT Service Management: Context-aware Knowledge Support for Incident Management
作者: 周柏村
Po-Tsun Chou
劉敦仁
Duen-Ren Liu
資訊管理研究所
關鍵字: 資訊科技服務管理;ITIL;資料探勘;情境認知;事件管理;ITSM;ITIL;Data Mining;Context-aware;Incident Management
公開日期: 2004
摘要: IT服務管理(Information Technology Service Management, ITSM),主要是將IT視為一種服務,運用服務管理的概念管理企業內的IT,藉由提供高品質的IT服務,提昇企業的競爭優勢。ITIL提供了一套IT服務管理的實務典範,以流程化的方式,進行IT服務管理。其中,事件管理流程負責處理日常營運所發生的突發事件,目的在於迅速地將事件解決或者提供暫時性解決方案(Workaround),以降低對業務運作的負面影響。ITIL所定義的事件管理流程中,只針對流程與步驟,以及應該注意的方針及要點,提出實務指引,並未探討如何透過適當的知識支援,以協助事件管理,增進事件處理效率。此外,事件通常有不同的情境,情境資訊的取得,可以協助處理人員掌握事件資訊,事件的情境也會影響事件的處理與解決方法。本研究提出情境認知的事件管理知識支援架構,運用資料探勘技術從事件記錄中發掘出與情境相關的事件管理知識,透過規則推論的方式提供情境認知的事件知識支援,有效掌握事件相關資訊,以協助支援事件管理,增進事件處理效率。
Information Technology Service Management (ITSM) regards IT as a service to manage IT from service-oriented perspective, whereas enterprises can raise their competence through high quality IT Services. ITIL has defined several processes of managing IT services to provide the guidelines for the best practices of ITSM. The Incident Management process handles incidents arisen from regular IT operations to reduce their negative effects on enterprise operations by providing first-aid resolutions or workaround solutions. ITIL-defined Incident Management process focuses on process activities, procedures, and guiding principles to provide practical instructions for handling incidents. Effective knowledge support for Incident Management has not been investigated. Furthermore, identifying the contextual information of incidents can provide more effective incident management by resolving the incidents according to their contextual information. This work proposes a framework of context-aware knowledge support for Incident Management. Association rule mining is employed to discover contextual knowledge of incidents from historical incident logs. Moreover, rule inferences are employed to infer possible contextual knowledge and resolutions. Consequently, effective knowledge supports for incident management can be achieved by providing needed contextual knowledge and possible resolutions through the rule mining and inferences.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009234529
http://hdl.handle.net/11536/77178
Appears in Collections:Thesis


Files in This Item:

  1. 452901.pdf