A methodology for modelling and designing robust information systems in an uncertain use environment

Typical modelling techniques for information system analysis and design treat key system requirements parameters as deterministic. In addition, system dynamics reflected in time-path behaviour, such as queues and bottlenecks, are not captured in traditional information process models; thus it is difficult to gain insights into the performance of an information system (IS) within the context of the uncertainty in its use environment. This is increasingly important for systems such as supply chain management and customer relationship management systems that operate within a global milieu. In this study we propose an experiment design methodology that, unlike conventional simulation experimentation, allows robust information system designs to be identified. We demonstrate the risks associated with design decisions in which uncertainty is ignored and provide results that illustrate how the proposed experiment design methodology reduces the ad hoc character of simulation experimentation and provides insights into the best design alternatives in light of uncertainty in the IS use environment.