A collective artefact design of decision support systems: design science research perspective

Purpose –The knowledge of artefact design in design science research can have an important application in the improvement of decision support systems (DSS) development research. Recent DSS literature has identified a significant need to develop user-centric DSS method for greater relevance with respect to context of use. To address this, this study develops a collective DSS design artefact as method in a practical industry context. Design/methodology/approach – Under the influence of goal-directed interaction design principles the study outlines the innovative DSS artefact based on design science methodology to deliver a cutting-edge decision support solution, which provides user-centric provisions through the use of design environment and ontology techniques. Findings –The DSS artefact as collective IT applications through the application of design science knowledge can effectively be designed to meet decision makers’ contextual needs in an agricultural industry context. Research limitations/implications – The study has limitations in that it was developed in a case study context and remains to be fully tested in a real business context. It is also assumed that the domain decisions can be parameterised and represented using a constraint programming language. Practical implications – We conclude that the DSS artefact design and this development successfully overcomes some of the limitations of traditional DSS such as low user uptake, system obsolescence, low returns on investment and a requirement for continual re-engineering effort. Social implications – The design artefact has the potential of increasing user uptake in an industry that has had relevancy problems with past DSS implementation and has experienced associated poor uptake. Originality/value –The design science paradigm provides structural guidance throughout the defined process, helping ensure fidelity both to best industry knowledge and to changing user contexts.