RAD-MRM: A Multi-Source Data-Integrated Methodology for Mining User Requirements in Industrial-Heritage Tourism

Abstract

To address limited sample coverage, fragmented requirement expressions, and weak strategy translation in industrial-heritage tourism, this study proposes the Real–Agent–Designer Multi-Source Requirements Mining methodology (RAD-MRM). It integrates 26 real-user semi-structured interviews, 26 Agent-user virtual interviews, input from three experience designers, 188 valid Kano questionnaires, and a 10-member evaluation panel. Kano analysis reduced 18 candidate requirements to 13 retained requirements. AHP identified intelligent route-flow management, experience retention, deep narrative, and real-time state sensing as high-priority requirements. FCE combined with entropy weighting showed that six experience-design strategies achieved an overall “Good” evaluation, with a comprehensive score of 3.264 and non-zero weights for all strategies. RAD-MRM therefore provides a structured pathway from multi-source requirement elicitation and screening to priority determination and strategy evaluation in complex cultural-tourism research and practice. Agent-user virtual interviews functioned only as a supplementary source of requirement expression and were not treated as substitutes for evidence from real users.