A data-driven four-dimensional evaluation framework for pediatric medical nebulizers: an integrated multi-criteria decision-making approach

To improve the scientific rigor and decision transparency of pediatric medical nebulizer design evaluation and to reduce the overreliance on experience-based judgment in scheme selection, this study proposes a four-dimensional comprehensive evaluation framework for pediatric medical nebulizers. First, real user needs were identified through online review mining and, together with findings from the literature review, market research, user interviews, and expert synthesis, were used to construct an evaluation index system covering four dimensions: visual, safety, interaction, and psychological factors. Second, subjective and objective weights were calculated using an improved analytic hierarchy process (AHP) and the CRITIC method, respectively, and these two types of weighting information were then coordinated through a game-theoretic combination weighting approach. Finally, an improved TOPSIS-RSR method was employed to rank and classify four conceptual design schemes for pediatric medical nebulizers. The results showed that indicators such as emotional guidance and relief, interactive playfulness, status monitoring, positive cognitive guidance, parental companionship and participation, and material biosafety carried relatively high comprehensive weights, with Scheme D demonstrating the best overall performance. Further sensitivity analysis based on weight perturbation indicated that the proposed framework exhibited good ranking stability under the present evaluation conditions. Overall, the proposed framework provides a more structured and relatively stable decision-making pathway for the evaluation of pediatric medical nebulizers. It can identify user needs, coordinate subjective and objective information, and provide more rigorous support for scheme selection. At the same time, it may also offer a useful methodological reference for other product design problems characterized by multi-source needs and multi-criteria trade-offs.