Applying a Variation of Delphi Method for Knowledge Elicitation in the Context of an Intelligent System Design

To appear in Advances in Human Factors and Systems Interaction , Series: Advances in Intelligent Systems and Computing
DOI:10.1007/978-3-030-20040-4_35

The effectiveness, ef?ciency and satisfaction are key indicators of the Usability of a system. The outcome of the knowledge management process associated with the development of intelligent systems is determinant in this regard. The paper describes the use of the Delphi method to identify the level of ?tness that subject matter experts (SMEs) recognize on different types of response units to perform speci?c disaster response tasks, with the purpose of expliciting the knowledge required for the inference process of the THEMIS intelligent system. Since the SMEs involved in this elicitation process had different levels of experience on different domains of disaster management, to help reaching consensus the Delphi method was modi?ed to weight the level of experience expressed by the SMEs. This variant of the Delphi method is presented, discussing the approach adopted in each step of the iterative process, the obtained results, and the followed validation process.