A framework for linking Soft Systems Methodology (SSM) and Jackson System Development (JSD)

Abstract

This paper describes a proposed framework for linking soft systems methodology (SSM), a problem structuring technique for use in messy, ill-defined problem situations, with Jackson System Development (JSD), an information system development methodology which has object-oriented characteristics. The approach taken has been to embed the modelling phase of JSD within SSM after the conceptual modelling stage, but before the debate on desirable and feasible change. The JSD modelling phase is carried out at a conceptual rather than a real world level. The framework comprises of a series of steps which enable JSD entity structure diagrams to be derived to a large extent from the activities in an SSM conceptual model. A data model may then be built up by exploding the activities in the conceptual model or by using 'conceptual hows' to identify more entities and actions. One of the main benefits expected to arise from the use of such a framework is an improved user requirement definition, which is important to successful systems implementation. The difficulties in actually achieving a link relate to the methodological and philosophical differences between SSM and JSD.

Untitled

“…One of the fundamental challenges of undertaking IS design is the need to find some means of moving from methods of inquiry suited to sense making in social situations, to methods suited to organizing knowledge into a suitable format for the construction of a logical specification for any supporting technology. Many authors have discussed the suitability of using Soft Systems Methodology (SSM) (Checkland, 1981;Scholes, 1990, 1999) to support initial sense making in a problem of concern, but problems have been encountered in linking up with methods for constructing technological specifications (Doyle, Wood, and Wood-Harper, 1993;Lai, 2000;Liang, West, and Stowell, 1998;Mathiassen and Nielsen, 2000;Miles, 1988;Prior, 1992;Savage and Mingers, 1996;Stowell and West, 1994). Most approaches to creating such a link have attempted to create some sort of bridge between the first phase of inquiry, exploration and sense making and the second phase, constructing a logical specification.…”

PEArL: a systems approach to demonstrating authenticity in information systems design

“…These models emphasize the organization of the data and messages that must be passed in order for the real world technological provision to function correctly. In the process of changing the emphasis from thinking about the action to be undertaken within the situation to a focus on creating a technical definition it is difficult to maintain a direct link between the two modelling approaches (Doyle and Wood, 1991;Stowell and West, 1994;Savage and Mingers, 1996). In order to enable clients to take ownership of the proposed change, there is a need for new modelling methods for supporting their complete participation (Stowell, 2000).…”

“…Others may be almost entirely excluded, particularly if methods requiring specific technical expertise are employed. For example, in an attempt to link SSM to Jackson's (1983) systems development, Savage and Mingers (1996) argued that the complexity of the notation might exclude some clients from participating in that phase of the design process. The manner in which people were engaged in the learning process is then a matter to reflect upon as fully as possible.…”

“…Many authors have argued that linking SSM to a method for deriving a technical definition for a computer-based IS would help overcome some of the disadvantages of technique-focused IS design (Stowell, 1985(Stowell, , 2000Sawyer, 1992;Lewis, 1993;Stowell and West, 1994;Mingers, 1995;Savage and Mingers, 1996;Liang et al, 1998). However, there are difficulties.…”

Untitled

“…Mingers (1988), in reply, acknowledged some similarities but indicated that there were fundamental philosophical differences between them; Avison et al (1998) also point out the philosophical difference and conclude that they are not similar. Other proposals include linking "SSM" to the Jackson System Development method (Savage and Mingers, 1996;Wood and Doyle, 1989) and re-expressing a model in a form that can be expressed in propositional calculus and predicate logic in order to derive a type of DFD (Gregory, 1993a-c;Merali, 1992).…”