Multi-Grounded Development of Enterprise Modeling Methods:The Case of SIMM
Abstract:
Multi-Grounded Theory (MGT) extends Grounded Theory by further grounding processes, most notably theoretical grounding.
As an example for applying MGT we use the enterprise modeling method SIMM, which is empirically grounded, and show that this method can profit from grounding it also in an external theory.
We outline a procedure called communicative and material functions analysis that can be used for this purpose and apply it to Business Action Theory.
With the help of the extended method we analyze a business situation in order to follow up the commitments that are made in the course of a business process with the ultimate aim of detecting flaws in that process.
Introduction:
Grounded Theory (GT; Glaser, Strauss 1967; Glaser 1992; Glaser 1998; Strauss, Corbin 1998) was introduced to provide a systematic procedure for deriving categories and theoretical constructs from empirical data.
GT is, strictly speaking, not a theory but rather a method for theory development.
Its characteristics are the relatively short iteration cycles and a process of continual improvement.
Starting with a small set of empirical data we develop an initial understanding of the problem field, which then can be used to gather further empirical data in a more controlled fashion, leading to an improved understanding and so on until we reach theoretical saturation.
GT started out in sociology but has since been applied in a variety of fields including informatics.
In the latter it has also been used to develop modeling methods, typically based on action research.
An example of such a method is SIMM (Situation-adaptable work and Information systems Modeling Method; Goldkuhl 1996).
SIMM has been empirically grounded in action research projects some of which are documented in (Lind, Goldkuhl 1997; Melin, Goldkuhl 1999; Axelsson et al. 2000; Goldkuhl, Melin 2001; Axelsson, Segerkvist 2001; Lind et al. 2003; Melin, Axelsson 2004; Johansson, Axelsson 2004; Johansson, Axelsson 2005; Haraldsson, Lind 2005).
SIMM provides both a method for enterprise modeling and a meta-method for evaluating modeling methods.
Empirical grounding proceeds as follows: SIMM Enterprise Modeling is applied in an action research project.
The experiences gained from the use of the method are evaluated with the SIMM Meta Method and corresponding changes to SIMM Enterprise Modeling are made.
In a further round of action research these changes are consolidated and further improvements are made (see fig. 1).
GT has been criticized for being restricted to a purely inductive approach.
It is argued that we cannot ignore the knowledge that established theories might contribute.
Consequently, (Goldkuhl, Cronholm 2003) suggest that further grounding processes are required in addtion to the existing empirical grounding process, most of all a theoretical grounding process. They call this new approach Multi-Grounded Theory (MGT).
If we want to apply MGT in the context of enterprise modeling we must identify a suitable theory that can function as external theory for theoretical grounding.
For the purpose of this paper we have chosen Business Action Theory (BAT; Goldkuhl 1996; Goldkuhl 1998; Goldkuhl, Lind 2004) because some preliminary work has already been done in that area. E.g. (Goldkuhl 1996) has extended SIMM to cover communicative actions, which are essential in BAT.
Many other important concepts such as layers and phases have been disregarded, though.
It is the purpose of this paper to complete the theoretical grounding of SIMM in BAT.
This can be considered as a first step, as a grouding in other action theories might also be considered.
But as the choice of external theory is contingent there are strong arguments to synthesize the relevant theories as Goldkuhl (2005) pointed out.
The result, Socio-Instrumental Pragmatism (SIP; Goldkuhl 2002; Goldkuhl 2005), is a general ontology of social action.
So far it is not sufficiently developed to provide a basis for theoretical grounding. We therefore restrict our approach to BAT for the purpose of this paper.
The remaining sections are structured as follows: We first introduce BAT and the generic layered patterns for business modeling.
This framework is then refined by combining phases and layers.
We proceed by applying communicative and material functions analysis to it, followed by a classification of the resulting functions, and finally leading to the extension of SIMM in terms of BAT concepts.
We conclude by presenting an application of the extended method for the purpose of commitment analysis.
Business Transaction Theory:
SIMM Extented with Business Action Theory:
Conclusion:
According to MGT a modeling method can be informed by grounding it in some theory or theories.
We have chosen Business Action Theory to inform the existing enterprise modeling method of SIMM.
Grounding is performed via a procedure called communicative and material functions analysis.
The result is an extended method that reflects the properties of the theory.
We have shown the usefulness of this approach in the context of a case study involving the analysis of commitments in an interorganizational business process.
Future research might investigate the theoretical grounding of enterprise modeling in a more general theory or ontology.