Process warehouses in practice: a goal‐driven method for business process analysis

Abstract

SUMMARY Process analysis and improvement, one of the core phases in the business process management life cycle, is reaching the top of the agenda for many enterprises and organizations. An emerging approach to analyzing business processes involves the use of process‐oriented data warehouses to collect information about executed processes to enable their detailed analysis and thereby provide bases for possible improvements. However, very little work has been carried out on employing business orientation in the design and utilization of a process warehouse. The approach presented in this study addresses that limitation by proposing a method that relates business goals to the information from the warehouse. To do so, we start by defining a desired goal structure that concerns an existing business process. The obtained goal model is firstly utilized for scoping and designing a corresponding data model for a process warehouse, and further for retrieving the information from the warehouse to examine the fulfillment of the goals of interest. The key benefits of the proposed method are the provision of the relevant and accurate business process information as well as tracing the process‐related goals to the process execution data and further to the analysis of the process. As validation of the proposed method, we develop a prototype, and a healthcare process is used as an illustrative example. Also, to demonstrate the method quality, we perform an experiment on the effectiveness of the information access and retrieval. Copyright © 2011 John Wiley & Sons, Ltd.

A Process Warehouse for Process Variants Analysis

“…[2] proposes to derive a generic data warehouse structures from the meta model of the BPMN, whereas [11] proposes a Sequence Warehouse (SeWA) architecture and OLAP tools to analyse data stemming from workflow logs but a conceptual model for DW is missing. Approaches based on goal-oriented methodology for requirement analysis in order to design a data warehouse were proposed by [10,15,19]. A multidimensional data modelling for business process analysis was proposed by [7,13,14,16] During our research work we have found a number of relational and multidimensional data warehouse design used for process mining analysis as well.…”