The field of complex product design evaluation can attract high ambiguity due to difficulties in establishing indicators and the subjectivity of expert evaluation scoring. Indeed, traditional Kansei Engineering (KE) relies on user requirements and feedback for design evaluation, which may not fully and effectively validate the design evaluation results, let alone determine whether they apply to complex products with more evaluation index systems. To overcome these drawbacks, this study proposes an evaluation method based on Hybrid Kansei Engineering (HKE) modeling for complex product design evaluation. HKE modeling consists of two parts, namely Forward Kansei Engineering (FKE) and Backward Kansei Engineering (BKE). In this study, four electric forklift designs are used as an example. The FKE system adopts the multi-attribute decision evaluation method; obtains the evaluation indexes of the forklift product imagery and then establishes the perceptual word collection; constructs the evaluation index system of the forklift via the Analytic Hierarchy Process (AHP); calculates the weights of the evaluation indexes of each level and their rankings; and calculates the final rankings of the four electric forklift design solutions by adopting the Fuzzy Comprehensive Evaluation (FCE) method. The BKE system adopts eye tracking (ET) to extract the attention time, visual attention hotspot, and other eye movement index data, and the Gray Relation Analysis (GRA) method was used to validate the model to derive the ranking, which verifies the effectiveness and scientific validity of the evaluation method. The results of this study show that the two-way evaluation of HKE modeling can effectively avoid subjective factors in product design, improve the scientific nature of the design, and guarantee the logical rigor of the perceptual design procedure.

User requirements are the core driving force behind the iterative development of complex products. Their comprehensive collection, accurate interpretation, and effective integration directly affect design outcomes. However, current practices often depend heavily on single-source data and designer intuition, resulting in incomplete, biased, and fragile design decisions. Moreover, multi-source heterogeneous user requirements often exhibit inherent asymmetry and imbalance in both structure and contribution. To address these issues, this study proposes a symmetric and balanced optimization method for multi-source heterogeneous user requirements in complex product design. Multiple acquisition and analysis approaches are integrated to mitigate the limitations of single-source data by fusing complementary information and enabling balanced decision-making. Firstly, unstructured text data from online reviews are used to extract initial user requirements, and a topic analysis method is applied for modeling and clustering. Secondly, user interviews are analyzed using a fuzzy satisfaction analysis, while eye-tracking experiments capture physiological behavior to support correlation analysis between internal preferences and external behavior. Finally, a cooperative game-based model is introduced to optimize conflicts among data sources, ensuring fairness in decision-making. The method was validated using a case study of oxygen concentrators. The findings demonstrate improvements in both decision robustness and requirement representation.

Background: As population aging and digitalization advance, smart interactive products are becoming part of older adults’ everyday lives. Although these technologies are expected to support independence, communication, and wellbeing, existing findings remain mixed. Prior research has focused mainly on technology acceptance or general use, with limited attention to how older adults’ psychological perceptions are translated into behavioral engagement and, in turn, subjective wellbeing. This study therefore proposes the Technology Engagement and Wellbeing (TEW) model to explain this mechanism. Methods: A questionnaire survey was conducted among adults aged 60 years and above with prior experience using smart interactive products. Of 2,686 collected questionnaires, 2,635 valid responses were retained. Psychological perceptions were represented by perceived usefulness, technology self-efficacy, and technology anxiety. Behavioral engagement was modeled as a multidimensional construct including usage frequency, functional diversity, and interactive depth. Structural equation modeling and bootstrap analysis were used to test the hypothesized relationships, and digital literacy was examined as a moderator. Results: The structural model showed good fit to the data ( χ ² /df = 2.618, CFI = 0.979, TLI = 0.977, RMSEA = 0.025). Psychological perceptions were positively associated with behavioral engagement ( β = 0.651, p < 0.001) and psychological need satisfaction ( β = 0.441, p < 0.001). Behavioral engagement was positively associated with psychological need satisfaction ( β = 0.255, p < 0.001) and subjective wellbeing ( β = 0.115, p < 0.001), while psychological need satisfaction was positively associated with subjective wellbeing ( β = 0.422, p < 0.001). The serial indirect association of psychological perceptions with subjective wellbeing through behavioral engagement and psychological need satisfaction was significant [ β = 0.070, 95% CI (0.058, 0.084)]. Digital literacy also significantly moderated the association between psychological perceptions and behavioral engagement ( β = 0.390, p < 0.001). Conclusion: The wellbeing value of smart interactive products lies not in access alone, but in whether positive psychological perceptions can be translated into active engagement that supports autonomy, competence, and relatedness. The TEW model provides an integrated framework for understanding how smart interactive product use becomes beneficial in later life.