Innovation of a design method (MoIST) that incorporates non-traditional 'soft' systems science into traditional 'hard' information systems design

The research explores ways of making the information systems development process
more effective. The thesis documents an original design method. This is the method
of incorporating Systems thinking into Information Systems Design (MoIST). The
thesis demonstrates that MoIST improves information systems design and adds to
the effective arsenal of methods that already exist.
The Computer Science literature has identified some weaknesses in the software
development methodologies. These weaknesses include premature design decisions
taken before major requirements are known. Another is the dearth of options for
applying Systems Science and information systems design techniques in a UMLbased
context. It was found that these weaknesses sometimes resulted in software
failures. These findings have been confirmed in the empirical and the evaluation
portion of the research.
The essence of the thesis is that appropriate software development strategies may
be chosen at various points in a project. The choice of strategy is based upon the
value of particular factors. These factors include confidence in requirements,
development environment structuredness, user types and developer types.
In order to achieve the research aims, the MoIST is utilised to preserve the
methodological strengths of the hard systems engineering paradigm. It
simultaneously attempts to minimise its weaknesses by combining it with a systems
science approach called Soft Systems Methodology (SSM). The research
incorporates this non-traditional ‘soft’ Systems Thinking into traditional ‘hard’
Information Systems Design. The two main contributions of the thesis are the
transformation of SSM conceptual models into UML use case diagrams and activity
diagrams. Another is the creation of MoIST Project Option Selection Tool (MoPros).
This MoIST method has been tested empirically by utilising it in a complex,
unstructured setting in a School of Computing and Engineering. Based on the
theoretical and practical work conducted, it is concluded that the MoIST method is
effective in several ways. It provides coherence and structure to complex software
projects and can help to facilitate decisions about improvement strategies. It also
successfully incorporates the results of SSM analysis into requirement specification based on the UML. The MoIST method is offered as a viable option to add to the
existing development alternatives for successful software development.