Identifying influential factors of business process performance using dependency analysis
作者: Branimir WetzsteinPhilipp LeitnerFlorian RosenbergSchahram DustdarFrank Leymann
作者单位: 1Institute of Architecture of Application Systems, University of Stuttgart , Stuttgart , Germany
2Distributed Systems Group, Vienna University of Technology , Vienna , Austria
3CSIRO ICT Centre , GPO Box 664 , Canberra , ACT , 2601 , Australia
刊名: Enterprise Information Systems, 2011, Vol.5 (1), pp.79-98
来源数据库: Taylor & Francis Journal
DOI: 10.1080/17517575.2010.493956
关键词: process performance monitoringservice compositionKPIQoSdata miningdecision tree
原始语种摘要: We present a comprehensive framework for identifying influential factors of business process performance. In particular, our approach combines monitoring of process events and Quality of Service (QoS) measurements with dependency analysis to effectively identify influential factors. The framework uses data mining techniques to construct tree structures to represent dependencies of a key performance indicator (KPI) on process and QoS metrics. These dependency trees allow business analysts to determine how process KPIs depend on lower-level process metrics and QoS characteristics of the IT infrastructure. The structure of the dependencies enables a drill-down analysis of single factors of influence to gain a deeper knowledge why certain KPI targets are not met.
全文获取路径: Taylor & Francis  (合作)
影响因子:9.256 (2012)