As the number of elderly patients with dementia is increasing year by year, accompanying products for the elderly with dementia are becoming more and more important. Senile dementia has become a major disease that affects countries around the world and, at the same time, it has brought enormous pressure on social care. Therefore, this research aims to obtain a new set of design methods for dementia elderly care products by introducing the combination of AHP and fuzzy evaluation methods and applying them to design examples. And then break through the existing design path of accompanying products for the elderly with dementia, and relieve the social burden. This study takes the intelligent care robot for the elderly with dementia as an example. Due to the particularity of user groups, this article proposed a design method for elderly care products that integrates mathematical theoretical models. This design method is convenient for designers to have a more accurate data reference when designing and positioning, and summarizing the supporting product design scheme that can better meet the actual needs of the elderly with dementia. Through SET factors to analyze the driving force of demand for dementia elderly care products, and combined with a questionnaire survey to establish the design requirements of dementia elderly care products for analysis of hierarchy process, it effectively defines the design demand attributes of dementia elderly intelligent care robots. The research results are applied to the design of the intelligent accompanying care robot for the demented elderly, the comprehensive fuzzy evaluation of the design scheme is carried out, and the direction of further optimization of the scheme is obtained. This article proposed a design method for accompanying care products for the elderly with dementia by combining F-AHP (Fuzzy Comprehensive Evaluation Method and Analytic Hierarchy Process). This design method is based on a rigorous mathematical theoretical model. Compared with the traditional design method, this research focused on quantitative analysis and evaluation of the actual design requirements of the dementia elderly care products and had more accurate design priority data reference in the scheme design. The research results show that based on the combination of F-AHP, this article redefines the design path of accompanying products for the elderly with dementia, and concludes that the functions that need to be improved and optimized in the future of intelligent accompanying robots for the elderly with dementia are mainly: executive ability training, computing ability training, and observation ability training. At the same time, the design process of the intelligent escort robot for the elderly with dementia can also be applied to the design of intelligent elderly products, which can improve the service experience of intelligent elderly products in a targeted manner.