Breaking the barriers of technology adoption: Explainable AI for requirement analysis and technology design in smart farming

Understanding the factors that drive and hinder technology adoption is critical for companies that try to access customer segments or governmental agencies that want to foster economic, ecological, or social change. By assessing the technological readiness of customer groups, common and individual barriers or opportunities for technology adoption can be observed and translated into technological requirements, business strategies, or policy interventions. Current approaches to assessing such barriers do not provide information on which factors influence technological readiness more than others, limiting the prioritization of targeted technological or political interventions. This research introduces an Explainable Machine Learning (XAI) approach to overcome this limitation. It exemplifies its usability for the Precision Livestock Farming domain, particularly for smart technologies incorporating novel advances in Artificial Intelligence and Internet of Things. A random forest machine learning model is introduced to identify clusters of different farmers' technological readiness based on the available features (survey questions). XAI techniques are then deployed to understand the influence of individual features on the prediction outcome, highlighting factors that increase or decrease technological readiness of farmers. The results are assessed for their potential for requirement and business analysis while providing targeted suggestions for technology design. • A technological framework for dynamic requirement and market analysis is proposed. • Non-linear dependencies in farm traits and technological readiness are modeled. • Drivers and barriers to Precision Livestock Farming technology adoption are analyzed. • Key factors influencing farmers' technological readiness are highlighted. • Data-driven strategies for tailored technology design are proposed.