基於本體論與風險驅動觀點軟體流程調適之研究;A Study of Ontology-based and Risk-driven Perspective on Software Process Tailoring

NCUIR > School of Management at National Central University > Graduate Institute of Information Management > Electronic Thesis & Dissertation > Item 987654321/67575

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/67575

Title:	基於本體論與風險驅動觀點軟體流程調適之研究;A Study of Ontology-based and Risk-driven Perspective on Software Process Tailoring
Authors:	蔡文隆;Tsai,Wen-lung
Contributors:	資訊管理學系
Keywords:	本體論;風險因子;軟體專案;軟體流程調適;專案知識;專家系統;Ontology;Risk factor;Software project;Process tailoring;Project knowledge;Expert System
Date:	2015-06-15
Issue Date:	2015-07-30 22:24:32 (UTC+8)
Publisher:	國立中央大學
Abstract:	對於軟體開發管理而言，組織通常會建立流程標準來讓各軟體專案據以執行。但鑒於軟體開發的獨特性，並非一套標準流程就可以完全套用在所有的軟體專案上。因此，企業組織需要訂定一套流程調適準則，讓專案成員根據不同專案所面臨的風險，依據調適準則而去調整實際的作法。然而，軟體流程調適是一個知識密集的活動，調適準則的制定或引用乃需要組織經驗方能提供適切的調適建議。故本文嘗試建立一個以規則為基底的決策支援系統，運用本體論(Ontology)來建置一個知識與風險觀點導向的軟體流程本體模型，並據以發展出軟體專案流程調適推薦系統(Ontology-based Software Process Tailoring Recommendation System簡為OntoSPRS)。本系統透過推論規則的設計，收集調適經驗，並以推論引擎對知識本體進行推論並產生調適策略，可協助專案成員在軟體開發過程中能更完整地將組織經驗帶入而導出具體實作方方式。本文並以一國內物流公司雲端企業資源規劃系統之實際案例來展示本文所建構的軟體流程本體。;To manage the quality and consistency of software development, many software organi-zations standardize the development processes for the software projects to follow. Owing to the uniqueness in software development, a process standard may not be fully applicable to all software projects. Therefore, organizations often allow projects to tailor defined processes from the standard based on ad-hoc project characteristics. However, software process tailoring is knowledge-intensive, and it needs organizational experiences in planning suitable processes for projects. In this regard, this paper attempts to develop a rule-based decision support sys-tem to serve this need. The system, called Ontology-based Software Process Tailoring Rec-ommendation System (briefly as OntoSPRS), includes a knowledge-based and risk-driven software process ontology model and features a rule inference engine for deriving tailoring strategies. To demonstrate the proposed work, a cloud-ERP project is presented. Findings as well as limitations of this study are also discussed.
Appears in Collections:	[Graduate Institute of Information Management] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	540	View/Open

社群 sharing

Loading...