EnOntoModel: A Semantically-Enriched Model for Ontologies

Introduction

Buzzwords such as BAM (Business Activity Monitoring), BOM (Business Operations Management), BPI (Business Process Intelligence) illustrate the interest in closing the business process management loop (van der Aalst and van Hee, 2002; Dumas et al., 2005). This is illustrated by the Figure 1, which shows the increasing level of support for closing the so-called BPM lifecycle.

Figure 1.

The level of support is rising - closing the business process management (BPM) cycle

The lifecycle identifies four different phases: process design (i.e., making a workflow schema), system configuration (i.e., getting a system to support the designed process), process enactment (i.e., the actual execution of the process using the system), and diagnosis (i.e., extracting knowledge from the process as it has been executed). BPM technology (e.g., workflow management systems) started with a focus on getting the system to work (i.e., the system configuration phase). Since the early nineties BPM technology matured and more emphasis was put on supporting the process design and process enactment phases in a better way. Now most vendors are trying to close the BPM lifecycle by adding diagnosis functionality.

The diagnosis phase assumes that data is collected in the enactment phase. Most information systems provide some kind of event log (also referred to as transaction log or audit trail). Typically such an event log registers the start and/or completion of activities. Every event refers to a case (i.e., process instance) and an activity (i.e., the step in the process executed), and, in most systems, also a timestamp, a performer, and some additional data. Process mining techniques (van der Aalst et al., 2003; van der Aalst et al., 2004; Agrawal et al., 1998; Cook and Wolf, 1998; Herbst, 2000; de Medeiros et al., 2003; Weijters and van der Aalst, 2003) take an event log as a starting point to extract knowledge, e.g., a model of the organization or the process. In the context of our ProM tool (van der Aalst et al., 2007) we are able to extract different types of process models (e.g., Petri nets, event-driven process chains, and instance graphs), social networks, organizational models, etc.

Existing techniques for process mining assume an event log to be in place. For many process-aware information systems (Dumas et al., 2005) this assumption is valid. For example, Workflow Management (WFM) systems, Enterprise Resource Planning (ERP) systems, Customer Relationship Management (CRM), Case Handling (CH) and Product Data Management (PDM) systems log information in some transaction log or audit trail. New legislation such as the Sarbanes-Oxley (SOX) Act (Sarbanes and Oxley, 2002) and increased emphasis on corporate governance have triggered the need for improved auditing systems (Hoffman, 2004). To audit an organization, business activities need to be monitored. As enterprises become increasingly automated, a tight coupling between auditing systems and the information systems supporting the operational processes becomes more important. However, many business processes are not directly supported by some process-aware information system. For many work processes relatively simple tools such as an e-mail program and text editor are being used. E-mail can be seen as the most popular tool used for Computer Supported Cooperative Work (CSCW) (Ellis, 2000; Ellis et al., 1991; Ellis and Nutt, 1996). The CSCW domain provides a very broad range of systems that support “work” in all its forms. WFM systems and other process-aware information systems can be seen as particular CSCW systems aiming at well-structured office processes. Therefore, it is worthwhile to explore the application of process mining in the broader CSCW domain. In this paper, we focus on e-mail systems and their logs.