The Construction of Critical Factors for Successfully Introducing Chatbots into Mental Health Services in the Army: Using a Hybrid MCDM Approach

Abstract

Previous research has shown that although military personnel are at high risk of developing mental disorders because of the excessive stress caused by their work, they also display low levels of intention to seek assistance because of the military culture. This, in turn, creates exorbitant costs for their respective countries. With the rapid development of artificial intelligence (AI)-related digital technologies, chatbots have been successfully applied to mental health services. Although the introduction of chatbots into the military to assist with mental health services is not common, it may become a future trend. This study aims to construct the critical factors for introducing chatbots into mental health services in the military, the relationships between the effects, and a weighting system, to ensure that the introduction of chatbots complies with sustainable practices. This includes four stages. In the initial stage, in accordance with the AI-readiness framework, in combination with the findings of previous research and specialist recommendations, preliminary indicators and items were developed. In the second stage, Fuzzy Delphi was used to confirm each dimension and indicator. In the third stage, using DEMATEL, an influential-network-relation map (INRM) of dimensions and indicators was created. In the fourth stage, using DANP, the relationships between the effects of the indicators and the weighting system were established. The findings of this study indicated that: (1) the key to success includes four dimensions and twenty-one indicators; (2) there is an interdependent relationship between the four dimensions and twenty-one indicators, and they influence each other; and (3) the four dimensions are technologies, goals, boundaries, and activities, in order of importance. Finally, specific suggestions are put forward to provide references for future practical applications and research, drawing on the results of this research.