Introduction

Under the current competitive market environment, precise design for specific target groups has become a key strategy for enterprises to realize long-term development1. In recent years, vulnerable groups such as children, the female and women have become hot spots for market segmentation due to their clear needs but low market saturation2,3,4. Among them, women, as a group that cannot be ignored in the society, the improvement of their consumption ability and social status has promoted the diversification of their market demand5,6,7,8,9. Especially in the medical field, female practitioners (e.g., healthcare workers, researchers, and administrators) account for a significant proportion of the population, and their occupational needs and physiological characteristics have put forward higher requirements for medical equipment design10. However, current medical device design still has systematic deficiencies: under the inherent male-centered design paradigm, female users are forced to adapt to standardized products based on male data, resulting in insufficient device size, comfort, and functionality, which in turn affects work efficiency, health and safety, and even the willingness to make career choices11,12,13,14,15.

Despite the attention paid to user needs analysis and product usability research16,17, there are still two limitations of existing methods:

  1. 1.

    Single dimension of user categorization: traditional studies mostly rely on demographic characteristics (e.g., age, occupation) to divide user groups, ignoring multidimensional differences in needs such as economic level, psychological preferences, and cultural values18,19,20. For example, although the literature21 focuses on the needs of healthcare workers for germ resistance and waterproofness, it does not analyze in depth the aesthetic needs of ordinary female users for device size and operating comfort.

  2. 2.

    Rationality assumption limitation of decision-making models: Traditional decision-making models such as expected utility theory EUT22,23assume that users are completely rational, ignoring the actual impact of cognitive limitations such as information overload or decision fatigue on the prioritization of design elements. For example, the literature24 optimizes dynamic product mix strategies through EUT, but fails to explain why users sacrifice some features in their actual choices.

In addition, with the rise of the C2B (Customer to Business) business model, diversified small-volume production has become the market norm25. This trend has exacerbated the disadvantages of the “one-size-fits-all” design model, and the contradiction between the objective gender differences and the solidified system of design elements has become more and more obvious26,27,28. In male-dominated design environments, the issue of gender-differentiated design has not been fully emphasized, resulting in the design of most products based on male needs and physiological characteristics, ignoring the special needs of female users26,29.

To solve the above problems, this study introduces the finite rationality assumption30,31and combines Pearson correlation analysis and hierarchical analysis method (AHP)32,33 to construct a multidimensional user classification framework. The framework breaks through the traditional method’s reliance on a single demographic characteristic, and provides more targeted design solutions by accurately identifying the differentiated priorities of different user groups for design elements. Specifically:

Multi-dimensional user classification framework: For the first time, the finite rationality assumption is combined with multi-dimensional classification of economic, knowledge, value, and psychological characteristics, and the differentiated priorities of female users for design elements are precisely identified through the synergistic application of Pearson correlation analysis and AHP method.

First, a multidimensional user classification framework based on the assumption of finite rationality is constructed, which breaks through the limitation of relying on a single demographic characteristic for user classification in traditional product design, and realizes the precise matching between user needs and design element priorities through the synergistic application of Pearson correlation analysis and AHP method; and through the synergistic application of Pearson correlation analysis and AHP method, realizes the precise matching between user needs and design element The precise matching of user demand and design element priority is realized through the synergistic application of Pearson correlation analysis and AHP method. Second, the theory of finite rationality is introduced into the field of product design, which compensates for the insufficiency of the traditional decision-making model’s assumption of complete rationality by recognizing the influence of users’ cognitive constraints on design decisions, and more realistically simulates the user’s preference for trade-offs between functionality, aesthetics, and comfort; third, the cross-disciplinary fusion forms a set of structured user-driven design paradigms, which provide a scientific basis for solving the contradiction between demand and fitness for the methodology for design optimization of women’s products.

Through this innovative framework, this study aims to reveal the deep connection between female users’ needs and product fit design, provide scientific basis for solving the contradiction of “clear needs but insufficient product fit”, and take protective clothing as the object of study, and while improving product fit, we intend to accurately localize and divide the characteristics of the target user groups, and carry out special design for their characteristics. In order to improve the overall product suitability, it is proposed to accurately position and classify the characteristics of the target user group and carry out special design to realize the optimal experience of product suitability. Meanwhile, this study also provides new theoretical tools and methodological references for user-driven product development in other fields.

Research framework

Existing product usability design is mostly based on the ideal needs of fully rational users to configure, in fact, affected by perceptual bias, in the product selection, the user can only be based on personal subjective rational judgment to choose a satisfactory solution, which achieves the maximization of the benefits of the individual’s motivation to use the product, the user’s needs and psychological feedback, which varies greatly and is highly susceptible to the interference of external environments, and then in the alternative solutions Re-select the relative optimal solution, therefore, most users are unable to be completely rational in judging products34, which is different from the assumption of complete rationality of users in traditional decision-making models such as expected utility theory EUT. The finite rationality hypothesis30,31, on the other hand, lies between complete and incomplete rationality, combining subjectivity and rationality of dynamic change process under certain restrictions. At present, the limited rationality hypothesis is used by many scholars in the study of individual decision-making behavior. For example, Zhao Dongmei et al.35 studied consumer demand behavior from the perspective of a finite rational consumer, and Li Jian et al.36 analyzed the consumer return decision under the C2B model based on the consumer’s finite rational behavior. In addition, Wan Tao et al.37 used finite rationality to study the conflict and cooperation of science and technology innovation team members and so on. The feature categories delineated with this guidance can effectively explore the implicit abstract demand information while obtaining the surface concrete demand information.

Analyzing product usability design priorities is a key part of understanding user needs and motivations for use to improve product usability experience38,39. In recent years, the Analytic Hierarchy Process (AHP)32,33combines qualitative and quantitative analysis, and has been widely used in the fields of product styling design40,41,42,43, mechanical structure design44,45,46, and evaluation of its design solutions47,48,49,50. In this paper, based on the results of user characteristics classification, we construct an evaluation analysis system through hierarchical analysis to rank and evaluate the product fitness design priorities corresponding to different types of users with different characteristics, so as to further improve the product fitness experience.

In order to accurately identify the differences in the needs of different user groups, and also in order to provide strong support for optimizing product design, and ultimately achieve the goal of improving user experience. This paper proposes a product design prioritization research method based on the classification of user characteristics as shown in Fig. 1, which contains the following key steps:

  1. 1.

    Collecting information about product unsuitability by means of data query and in-depth interview, and combining with the theory of the finite rationality hypothesis, a detailed questionnaire is designed so as to comprehensively understand the users’ actual needs and usage experience.

  2. 2.

    Pearson’s correlation analysis was used to analyze the collected questionnaire data in depth, to understand the relationship between the demand intensity of different user groups for a product and their selection behaviors, in order to extract the main factors affecting the subjective attitude of users, so as to realize the effective classification of user group characteristics.

  3. 3.

    In order to further evaluate the usability of the product, the hierarchical analysis method is used to construct a product usability evaluation analysis system to identify the differences in the product selection preferences of different types of users. The comprehensive weights of product usability design priorities corresponding to each type of user are calculated to quantitatively rank product usability requirements, and design solutions are adjusted according to the priorities to ensure that a better product experience is provided to specific user groups.

Fig. 1
figure 1

Product design priority configuration method based on user characteristics classification.

Full size image

User feature classification method based on bounded rationality hypothesis

Due to the different characteristics and attributes of users, their motivation to use the product, user demands and psychological feedback will produce large variability, which in turn affects the intuitive preference of users for product use. Therefore, it is particularly important to scientifically and accurately develop a feature classification method. In this study, based on the progressive relationship of the depth of the demand hierarchy, the feature classification from the figurative material factors to the abstract perceptual factors is formulated, aiming to have the opportunity to explore the hidden depth of demand information while obtaining the superficial demand information through such a classification basis.

Considering that users’ needs and choices of products are decision-making behaviors after rational thinking34. Subject to the constraints of cultural level, information collection, and computational ability, the intensity of demand and product selection behavior shown by users is only a satisfactory decision made by combining internal and external factors to the external environment, the user’s decision-making on the product is limited rationality, and this decision-making changes accordingly with the changes of personal conditions and external environment.

Therefore, in order to classify user characteristics, we explored important factors related to usability experience that influence users’ product selection behavior. Based on the assumption of finite rationality, we used research methods such as data surveys, in-depth interviews, and expert consultation to extract the factors that shape the phenomenon of product unsuitability and establish a multidimensional analytical framework, as shown in Fig. 2. By considering these broader categories, we hope to gain a more nuanced understanding of user needs and preferences, thereby reducing the risk of drawing misleading conclusions from oversimplified individual attributes.

The study was conducted by means of an anonymous questionnaire, and the experimental protocol was approved by Yantai Nanshan University, which confirmed that the experimental protocol complied with the principles of the Declaration of Helsinki and the ethical exemption requirements of the Measures for Ethical Review of Life Science and Medical Research Involving Humans issued by China, and that the experimental protocol was exempted from ethical review. To ensure that participants were fully aware of the requirements for participation, the following aspects were explained: (1) the expected time for completion of the questionnaire; (2) the approximate scope of the questionnaire; and (3) the principles of voluntariness and confidentiality of the study. Prior to the start of the study, each participant is presented with a detailed informed consent form that lists the specific study procedures, potential risks and benefits, privacy measures, and the right to withdraw from the study at any time. After reading and understanding the consent form, the participant signs to confirm their agreement to participate in the study, ensuring that the study complies with ethical standards and legal requirements. All aspects of the study, from questionnaire design, distribution, and collection to data entry and analysis, follow the standardized academic research protocols to ensure the scientificity, rigor, reliability and validity of the study, and contribute to the further in-depth study of “Research on product design priority allocation based on user characteristics classification" provides a solid data foundation and methodological guarantee.

User characteristics classification methods

On the basis of the above analytical framework, we designed and distributed a comprehensive assessment questionnaire based on multiple dimensions to collect data on various aspects of Personal Factors, Unit Fac tors, Family Factors, Social Factors, Work Fac tors, and Cognitive Factors. We then extracted relevant factors using Pearson’s correlation coefficient to quantitatively analyze the important factors affecting product demand intensity and product selection behavior, and used this as the basis for user categorization, which ensured that user categorization was not based on a single self-reported measure alone.

The steps are as follows:

Fig. 2
figure 2

Analysis framework.

Full size image

Step 1: Determine the specific factor variables for each of the basic factors in the above analytical framework, with a total of 17 specific factor variables identified and coded as independent variables in the order of X1 to X17. The intensity of demand for the product is set as Y1 and the product selection behavior is set as Y2.

Step 2. Pearson correlation coefficients were calculated by substituting the designed independent and dependent variables into the following theoretical model as shown in Eq. (2) to test the factor variables for which significant effects occurred.

$$\begin{gathered} \rho ({\text{x,y)}} = \frac{{{\text{COV (}}X,Y)}}{{{\sigma _x}{\sigma _y}}} = \frac{{E((X - {\mu _x})(Y - {\mu _y}))}}{{{\sigma _x}{\sigma _y}}} = \frac{{E(XY) - E(X)E(Y)}}{{\sqrt {E({X^{\text{2}}}){\text{ - }}{E^2}(X)} \sqrt {E({Y^{\text{2}}}{\text{) - }}{E^2}(Y)} }} \hfill \\ \rho ({\text{x,y)}} = \frac{{N\sum {XY - \sum {X\sum Y } } }}{{\sqrt {N\sum {{{\text{X}}^{\text{2}}} - (\sum {X{)^2}} } } \sqrt {N\sum {{Y^2} - {{(\sum Y )}^2}} } }} \hfill \\ \hfill \\ \end{gathered}$$
(1)

Based on the above formula, the code is implemented using Python 3. The full code and related resources are hosted on GitHub and can be accessed here.

Step 3. The geometric model of the results against the Pearson correlation coefficient is shown in Fig. 3. When the correlation coefficient is 1, it becomes a complete positive correlation; when the correlation coefficient is −1, it becomes a complete negative correlation; the larger the absolute value of the correlation coefficient is, the stronger the correlation is; the closer the correlation coefficient is to 0, the weaker the correlation is. According to the variable correlation strength determination Tables51,52 shown in Table 1, the significance factors were selected to categorize the user group characteristics. In order to assess the statistical significance between user preferences, we introduced confidence intervals and p-values in the correlation analysis53, and we computed the Pearson correlation coefficients using the SciPy library in Python along with the corresponding 95% confidence intervals and p-values. For each correlation coefficient, the correlation was considered statistically significant if the p-value was less than 0.05. This approach ensured that we were able to accurately identify which user preferences had significant correlations with each other.

Fig. 3
figure 3

Geometric model of Pearson correlation coefficient.

Full size image
Table 1 Judgment table of variable correlation strength.
Full size table

Product design prioritization configuration method based on AHP approach

Design element preferences for products

We used an extensive literature review54,55,56,57 to identify the most frequently cited key factors in protective clothing design, such as comfort, functionality, and exterior design. In addition, we consulted several experts in the field and collected opinions from 137 female users through interviews. Based on this data, we shortlisted the most representative design elements, including flexibility, fit, color selection, and durability. After several rounds of internal discussions and external expert reviews, three main design variables were finally summarized: product attributes, product style and symbolic elements, and product human factors, and their subordinate nine secondary indicators. The selection of each variable is based on its importance in the existing literature and the actual needs of users’ feedback, which ensures the scientificity and rationality of variable selection.

The impact of these variables on user preference and satisfaction is well documented, and their inclusion is justified by prior research and empirical evidence40,42,47. The inclusion of each variable was validated through iterative consultations with end-users and industry professionals to ensure comprehensiveness and relevance. The AHP is a multi-criteria decision-making methodology that addresses the challenge of prioritizing product design elements. It involves more stakeholders in the decision-making process by comparing only two criteria, which can be transferred to an online questionnaire. Although most studies tend to involve experts in decision-making46,47. However, we believe that they are the experts who know themselves, and as Zhu et al.40 pointed out, it is feasible and effective to collect preference information directly from end-users for AHP analysis in some contexts. Therefore, we practiced the AHP method on users and conducted a subjective questionnaire survey on product usage preferences for a user group after characterization so as to explore the differences in the characteristics of product usage preferences among various types of female users and to lay the foundation for subsequent research on targeted design adjustments to improve the experience of this user group.

Construction of product suitability evaluation and analysis system

In order to solve the complex decision-making problems related to design elements in products58,59, a multi-level hierarchical product suitability evaluation and analysis system is established in Fig. 4, which is divided into three primary indicators and nine secondary indicators of product attributes, symbolic elements, and human-computer factors according to the differences in users’ preferences for product design elements. The system is divided into 3 primary indicators and 9 secondary indicators for product attributes, symbolic elements and human-machine factors. Hierarchical analysis is conducted for each user group, and the weight of each indicator on the user’s product experience is determined by comparing each other, i.e., the user experience judgment matrix is constructed to find out the priority ranking of the user group in the product design elements so as to make targeted design improvements.

Fig. 4
figure 4

Evaluation and analysis system for suitability of female protective products.

Full size image

Product Attribute Definition (B1).

Physiological Characteristic Differences (C1): Consider the physiological characteristics of different user groups, such as height, arm span, hand size, etc., to ensure that the product design can be adapted to a wide range of user groups.

Gender-specific Selling Points (C2): Design specific features or appearances for different genders, e.g., female users may prefer soft colors and small designs, while male users may prefer sturdy and functional designs.

Potential Gender Users (C3): Designing with the gender preferences of potential users in mind ensures that the product appeals to a wide range of people and avoids gender stereotyping.

Product Symbol Element (B2).

Shape Element (C4): The overall shape and structure design of the product, including streamlined design, geometric shapes, etc., to ensure aesthetics and practicality.

Color (C5): The selection and matching of colors affect the visual appeal and emotional expression of the product.

Material Element (C6): The choice of material and texture, affecting the touch and durability of the product.

Product Ergonomics Factor (B3).

Physiological Scale (C7): The size and proportion of the product to ensure conformity with the principle of ergonomics, so that the user can feel comfortable in the process of using the product.

User Interface (C8): The design of the interactive interface, including button layout, screen display, etc., ensures that users can easily understand and use it.

Operating Space (C9): Whether the user has enough physical space to operate the product comfortably and efficiently while using it.

Definition of scale of judgment matrix scaling

Construct m pairs of \(n*n\) pairwise comparison matrices. Denote the result of the comparison of the ith factor relative to the jth factor by \({a_{ij}}\), which indicates the degree of influence of each factor in the layer on a factor in the previous layer, and determines the weighting in the layer relative to a given criterion, and the nine levels of scaling taken for the comparisons are shown in Table 2.

Hierarchical single ordering and consistency tests

Hierarchical single ranking32,33 is based on the judgment matrix to calculate the importance weights of the indicators linked to the hierarchy for the indicators of the previous level. Calculate the eigenvalues and eigenvectors of the judgment matrix, i.e., calculate the eigenvalues and eigenvectors of the judgment matrix that satisfy the following relationships:

$$DV = {\lambda _{{\text{max}}}}V$$
(2)
Table 2 Nine-level scaling method.
Full size table

In Eq. 3, \({\lambda _{{\text{max}}}}V\) is the largest eigenroot of the judgment matrix D, and V is the eigenvector corresponding to \({\lambda _{{\text{max}}}}V\). In order to ensure the quality of the judgment matrix, its consistency is checked. Each index value in the judgment matrix should satisfy:

$${{\text{b}}_{{\text{ij}}}} = \frac{{{{\text{b}}_{{\text{ik}}}}}}{{{{\text{b}}_{{\text{kj}}}}}}{\text{(i,j,k}} = {\text{1,2,}} \ldots \ldots ,{\text{n)}}$$
(3)

Here, \({{\text{b}}_{{\text{ij}}}}\) represents the importance ratio of the the i th element to the j th element, while \({{\text{b}}_{{\text{ik}}}}\) and \({{\text{b}}_{{\text{kj}}}}\) represent the importance ratio of the i th element to the k th element and the k th element to the j th element, respectively. With this formula, we are able to indirectly calculate the relative importance between any two elements, which helps to test and improve the consistency of the judgment matrix.

In the actual analysis, due to the differences in the complexity of objective things, so that each judgment matrix has full consistency is impossible, in order to examine whether the judgment matrix can be applied to the hierarchical analysis, we have to do the consistency test of the judgment matrix. In order to test the consistency of the judgment matrix, it is necessary to calculate the consistency index:

$$CI = \frac{{({\lambda _{\max }} - n)}}{{(n - 1)}}$$
(4)

In Eq. 7,n is the order of the judgment matrix. The consistency check algorithm is better when the order of the matrix is less than 3. When the order of the matrix is higher, the consistency has to be corrected. The operator is as follows:

$$CR = CI/RI$$
(5)

RI is the correction factor and the values are shown in Table 3:

Table 3 Correction factor value table.
Full size table

When \(CR<0.1\), then the matrix satisfies the consistency requirement.

Product fit design prioritization Weights Calculation

Calculate the target criterion layer weight vector as:

$$V = ({{\text{v}}_1},{{\text{v}}_2}{\text{,}}{{\text{v}}_3}{\text{,}} \ldots \ldots ,{{\text{v}}_{\text{k}}})$$
(6)

Vi is the relative weight of criterion level indicator i in the criterion level.

Let for the kth criterion level indicator, the weight of the measure level indicator below each criterion be:

$${V_{i,j}} = {V_i} \cdot {V_j}$$
(7)

Then the operator for calculating the combined weights of the indicators for measure j under criterion i in the hierarchy is:

The combined weights of each indicator of product design prioritization are calculated and ranked in order of magnitude to obtain the importance results.

Product design prioritization configuration

The sensitivity analysis’s helps to assess the impact of changes in different variables on the prioritization and the credibility of the study’s conclusions by varying the weights (± 10%) of each design variable to simulate changes in the prioritization under different scenarios. After each weight adjustment, the consistency ratio (CR value) of the judgment matrix was recalculated to ensure that it was still less than 0.1, indicating that the adjusted judgment matrix maintained good consistency. Based on the adjusted weights, the priority of each design variable was recalculated and compared with the original priority to assess its stability.

Product design prioritization configuration

According to the comprehensive weight ranking of product design priorities, the design elements with high comprehensive priority are reasonably selected for improvement and enhancement. This method makes up for the limitation of using a single design requirement and design strategy to serve all users’ thinking60, and truly realizes the shared experience of the product.

Case studies

Prioritization of product design for women’s protection

Taking female protective products as an example, considering the wide range and necessity of protective clothing as a representative protective product applied in high-risk operating environments61,62,63,64, it is proposed to select female protective clothing as a research carrier, and apply the method of this paper to carry out the priority configuration of product design based on the categorization of user characteristics, and to verify the validity and reliability of the method.

The questionnaire was constructed based on the analytical framework in Fig. 2, and it was distributed to female users to collect data. To ensure the content validity of the questionnaire, relevant professionals reviewed and modified it multiple times during the development phase. For the reliability test, we used the Cronbach’s Alpha coefficient in SPSS to assess the internal consistency of the questionnaire. The calculated results show that the Cronbach’s Alpha value of the overall questionnaire is 0.732, indicating that the questionnaire has good internal consistency, thus supporting the feasibility of the finite rationality hypothesis model. The target respondents included frontline healthcare worker, community volunteer, research and logistics staff, and housewive. A total of 260 questionnaires were distributed between March 1, 2024, and March 20, 2024, with 235 valid responses received. The questionnaire recovery rate was 90.38%. The high recall rate not only indicates the participants’ strong interest in the survey topic, but also enhances the representativeness and credibility of the findings.

The results of the questionnaire were analyzed by running the Pearson’s correlation coefficient code set up in Python 3. The results are shown in Table 4.

Table 4 Questionnaire result statistics.
Full size table

Considering only factors with statistically significant correlation (p < 0.05), we classify users into different categories based on their preferences and needs. Empirical research shows that: on the one hand, due to the strengthening of the economic characteristics and knowledge characteristics of female users, they have a stronger desire to protect their rights and interests; at the same time, policies and other institutional requirements are stringent, the concept of epidemic prevention, power consciousness and other cognitive levels of the gradual improvement will also show a higher intensity of demand for the possession of shared protective products; on the other hand, it is also mainly the economic characteristics, knowledge characteristics, the concept of epidemic prevention and other factors that On the other hand, it is mainly the economic characteristics, knowledge characteristics, concept of epidemic prevention and other factors that prevent female users from choosing shared protective products. Following the above significant factors (economic characteristics, knowledge characteristics, value characteristics, psychological characteristics) that affect the user needs and motivation of female user groups, female users are divided into four categories according to the intensity of their needs, as shown in Fig. 5, to explore the sensitive preferences of female users for protective products, i.e., to prioritize the elements of the design of protective products for female users, in order to improve the suitability of the products.

Fig. 5
figure 5

Classification of female user characteristics.

Full size image

The classification of female users’ characteristics is shown in Fig. 5:

  • - Group A (n = 43) female users are in an introverted rectangle in the four-quadrant coordinate system, indicating they are at a disadvantage in all four characteristics.

  • - Group B (n = 67) female users have moderate economic and knowledge characteristics, along with greater value and psychological characteristics.

  • - Group C (n = 55) female users also have moderate economic and knowledge characteristics, but their value characteristics are somewhat lacking.

  • - Group D (n = 70) female users show an outward rectangle in the four-quadrant coordinate system, indicating they possess high quality in all four characteristics.

On this basis, we further applied demographic and socio-economic indicators to carry out detailed statistics on the user groups of subjects categorized by characteristics. The statistics of age, income, education level and family structure of female users are shown in Table 5.The results show that the average age of female users in Class A and B is much higher than that of Class C and D, with Class C having the smallest average age; the average education level of Class C and D is higher than that of the other two groups; the average income level of Class D is the highest, and the average income level of Class A is the lowest. It can be seen that it is inaccurate to categorize female users only in terms of age, income, education, family structure, etc., and the categorization factors are diverse and comprehensive. Therefore, classification based on the characteristics of female users is not only necessary but also helps to clarify the priority of product design elements for a certain user group.

Table 5 Statistics of age, income, education level andfamily structure of female users (unit: person).
Full size table

Combined with the questionnaire on the four groups of female users product selection preference subjective survey results are shown in Fig. 6, it is proved that female users in the product attributes, symbols, and human-computer factors of the three major indicators of the product design elements of the preference of the larger, for the subsequent study of the needs of female users to lay the foundation.

According to Fig. 4, the four types of female user groups are analyzed hierarchically, and the weights of the user interface factors on the product experience of female users are determined by comparing them with each other, i.e., the user experience judgment matrix is constructed. In order to quantitatively display the priority of each element in the matrix, a matrix judgment scale (1–9 scale method) is used, see Table 2, to construct the judgment matrix, and SPSS is used to calculate the comprehensive weights of the priority of each design element in the matrix. Taking the judgment matrix for female users of category D as an example, the judgment matrix is constructed by taking the common use experience of female user protection product A, product attribute definition B1, product symbol element B2, and product man-machine B3 as the criterion layer. The comparison results of each element are shown in Table 6. It can be found that type D female users pay more attention to the common use experience of product symbol elements, in which the user experience of product design elements such as physiological scale, color elements, and material elements have higher priority. This is different from the prioritization of protective clothing design elements derived from protective clothing suitable for the general public in literature15, and also different from the prioritization of protective clothing design elements derived from protective clothing for a wide range of female users in literature21.

Fig. 6
figure 6

Differences in product selection preferences among four types of female user groups.

Full size image
Table 6 Judgment matrix and results.
Full size table

In order to verify the reliability and stability of the constructed AHP model, a sensitivity analysis was performed. Specifically, for each two-by-two comparison matrix of Product suitability experience (A), Product Attribute Definition (B1), Product Symbol Elements (B2), and Product Man-Machine Factors (B3), the weights of their internal elements were adjusted respectively, where the original weights of B1 were adjusted from 0.05 to 0.0719, and the original weights of B1 were adjusted from 0.05 to 0.0719, respectively. Comparison matrix, respectively, to adjust the weights of its internal elements, in which the original weight of B1 is adjusted from 0.0719 to 0.05 or 0.1, the original weight of B2 is adjusted from 0.2796 to 0.25 or 0.3, and the original weight of B3 is adjusted from 0.6491 to 0.6 or 0.7. Observe the impact of these changes on the final decision-making results. 0.6491–0.6 and 0.7. Observe the impact of these changes on the final decision results. We find that the priority ranking of the physiological scales remains stable despite fluctuations in the weights; however, the priorities of the color and material elements change slightly at some extremes, but the overall trend remains unchanged. This suggests that although the color and material elements are of some importance in the minds of users, the physiological scale and workspace have become more critical in actual product design. Then, the consistency index (C.I.), random consistency index (R.I.), and consistency ratio (C.R.) were calculated for each matrix. The results show that the C.R. of all the matrices is less than the standard threshold of 0.1, proving that the two-by-two comparison matrices have a satisfactory level of consistency. This means that the portfolio weight values calculated based on these matrices are reliable and can be used in the subsequent decision-making process.

Table 7 Matrix comprehensive weight ranking of four types of female feature users.
Full size table

Based on Eq. 4 and Eq. 5 for the consistency test of the resulting matrix, the test qualified four categories of female characteristics of the user matrix comprehensive weight ranking is shown in Table 7, it can be obtained: D, C, B categories of female users are on the physiological scale elements of high priority; C, A categories of female users on the color elements of high priority; the material elements of the priority of the weight is stable and flat, the four categories of female users are on it with higher priority.

In response to the above high-priority design elements for female users in category D - physiological scale, workspace, and material elements - the common medical protective clothing on the market has been improved and designed. The main difference between the two types of medical protective clothing is that the resulting product has newly evaluated and adjusted the size and proportion of the protective clothing to better adapt to the physical characteristics of female users, and added stretch pleats on the basis of the existing market protective clothing to provide more spacious operating space. The main differences between the two designs are the new evaluation and adjustment of the size and proportion of the gowns to better fit the female user’s body shape, the addition of stretch pleats at critical body lines to provide more room for maneuvering, high-quality and tactile materials, and waterproof and antimicrobial surfaces to the existing gowns. Satisfaction statistics for each of the two designs.

In order to ensure the authenticity and effectiveness of the test results data, this questionnaire combined with the design elements obtained in the previous section, using a 7-level Linket scale to test the user’s mental imagery, the test results are shown in Fig. 7. among them, the tested users see the design elements of the design of the picture of the design case scoring, 1 point indicates that the user is biased in favor of the left-hand side of the design scheme, and vice versa, 7 points indicates that the user thinks that the right-hand side of the design scheme is better. A total of 250 questionnaires were distributed, and 242 questionnaires were effectively recovered.

Fig. 7
figure 7

Protective Clothing Design Program Satisfaction Statistics.

Full size image
Fig. 8
figure 8

Satisfaction of four types of female users with protective clothing design solutions.

Full size image

The results of the statistical questionnaire using SPSS software are shown in Fig. 8. After improving the physiological scale limitations of medical protective clothing, adjusting the workspace, and improving the material, the product satisfaction of Option 2 was significantly improved among users of categories d and c. The satisfaction of users of category b was not obvious but improved, and the adjustment of the product design elements had almost no effect on female users of category a. The results show that different types of female users generally approve of the optimization of the fit experience of protective products. These findings emphasize the importance of targeted design modifications and confirm the feasibility of the product design prioritization research methodology proposed in this paper based on the classification of user characteristics.

Women’s protection product design prioritization configuration application

Through the above design case study and observation and analysis of Fig. 5; Tables 5 and 7, we summarize the priority characteristics of the experience of protective clothing for the four types of female users based on the characteristic factors (economic characteristics, knowledge characteristics, value characteristics, and psychological characteristics) that affect the user needs and motivation for use of the female user group. Female users in category A are disadvantaged in all aspects of their characteristics, have low demands for the function and use of protective products, and tend to accept protective products in a passive manner; the complexity of the user interface and other elements of the improved protective products is negatively correlated with the suitability of the protective products, and special instructions are needed.

  • Compared with Class A female users, Class B female users have a certain economic and knowledge base, and have higher expectations for life. The priority of the demand for the fitness of protective equipment is focused on the user interface of the product function display, and they have higher requirements for the functionality and ease of use of the fitness of protective equipment.

  • Class C women are in the middle level of all aspects of the characteristics of the physiological scale and other product human factors have a great demand, protective equipment fit needs and class B users are very similar;

  • Female users in category D have higher knowledge and economic base, have higher requirements for work quality and themselves, and have high demand for the suitability of protective equipment, so they can emphasize on re-adjustment of human-machine factors of the products; at the same time, they have considerable confidence in their cognitive ability, which means that they have great autonomy and exploratory power in the use and learning of new products.

By focusing on and prioritizing these elements, the needs of this type of female user can be.

better met, and the usability and user satisfaction of products for women’s protective apparel can be better targeted.

Discussion

In this study, the factors influencing the phenomenon of high demand and low difference in the use of female shared protection products are used as the basis for segmentation to accurately characterize and segment the target female users. As shown in Fig. 5, the four types of female users have some degree of characteristic differences in product selection preferences. Hierarchical analysis verified that the differences in preferences of different types of female users and the differences in the prioritization of design elements have a significant impact on the fitness of protection products. As shown in Table 7, in this study, different types of female users place different levels of importance on product design, which also reflects that different types of users place different levels of importance on product experience. Improving the design of protective clothing for the design elements of concern to female users in category D and conducting a satisfaction survey, the results show that the four types of female users have different levels of recognition, and the percentage of satisfaction improvement is the highest for female users in category D. This ranking confirms the feasibility of the research method.

The research method combines the finite rationality assumption, Pearson correlation analysis and AHP to form a systematic design optimization framework. Unlike the traditional “one-size-fits-all” strategy, the framework achieves precise targeting of product improvement through multi-dimensional user classification and prioritization of design elements. Considering that traditional decision-making models such as EUT22,23 usually assume that decision makers are completely rational and individuals always seek to maximize the expected utility when making decisions, whereas finite rationality theory takes into account the limitations of human cognitive resources, and believes that individuals will adopt simplified strategies to cope with complex environments during decision-making. The comparison reveals that the theory of finite rationality has higher accuracy and practicality in explaining the decision-making behavior of actual users, especially in the face of information overload and time pressure. Thus, the finite rationality assumptions we use provide a more realistic model for understanding how users prioritize design elements.

The present approach offers significant advantages in user classification and decision modeling. Compared to demographic-based user segmentation methods, the multidimensional classification method based on finite rationality is more capable of identifying the heterogeneity of needs within a group and capturing the individualized needs of the group in greater depth. In addition, we also compare and analyze with the commonly used user classification methods such as Cluster Analysis65,66, Factor Analysis67,68, and Social Network Analysis69, and we find that Cluster Analysis is more effective in identifying the heterogeneity within a group than demographic-based multidimensional classification methods. heterogeneity within a group, but its results depend on the choice of initial parameters, which may lead to different segmentation schemes. Factor analysis helps to simplify the complex data structure by reducing data dimensions to extract key variables, but its validity and reliability are limited by sample size and data quality. Social network analysis focuses on depicting interaction patterns among users and is suitable for understanding user behavior under social influence, but its application is limited in the absence of sufficient social data. In contrast, finite rationality-based methods are better able to capture the dynamics of users in the decision-making process, thus providing segmentation results with more stability and explanatory power. However, we also recognize that different approaches have their own advantages and disadvantages, and future research can further optimize user segmentation strategies by combining multiple approaches.

For decision modeling, we also compared the AHP method with the Analytical Method of Hierarchy (ANP)70,71and Fuzzy Set Theory (FST)72,73.ANP allows for interdependencies among the evaluation criteria, which gives it an advantage when dealing with complex multilevel decision problems. However, ANP has a high computational complexity and the results are difficult to interpret. On the other hand, fuzzy set theory deals with uncertainty and ambiguity by introducing fuzzy logic, which is suitable for decision-making environments with high uncertainty. However, fuzzy set theory requires more parameter setting and subjective judgment, which may lead to instability of the results. Considering that when choosing protective clothing, users may make decisions based on personal experience, immediate needs, and limited information, rather than strictly following the principle of maximizing expected utility. In contrast, the AHP method is not only easy to operate, but also easy to understand and interpret, making it particularly suitable for the user preference analysis involved in this study. In addition, the sensitivity analysis verified that the weight assignment could maintain the stability of the priorities under ± 10% fluctuation (CR < 0.1), further improving the reliability of the method.

Conclusions

This study reveals the correlation mechanism between female user needs and product fitness design through a systematic approach, and proposes three innovative contributions: first, a multidimensional user classification framework based on the assumption of finite rationality is constructed, which breaks through the limitation of relying on a single demographic characteristic for user classification in traditional product design, and through the synergistic application of Pearson correlation analysis and the AHP method, an accurate match between user needs and design element priorities is realized. Second, the theory of finite rationality is introduced into the field of product design, which makes up for the inadequacy of the complete rationality assumption of the traditional decision-making model by recognizing the influence of users’ cognitive constraints on design decisions and more realistically simulating the users’ preferences for trade-offs between functionality, aesthetics and comfort. Third, the interdisciplinary integration forms a set of structured user-driven design paradigms, which provides a scientific basis for solving the contradiction between demand and fitness, and its methodology can be directly applied to the design optimization of women’s products, avoiding the waste of resources caused by “one-size-fits-all” design.

However, there are still some limitations in this study: first, the data collection focuses on female users in a specific region, which fails to fully consider the influence of different cultural backgrounds on the prioritization of design elements, which may limit the generalizability of the methodology; second, the subjective evaluation of the users may be affected by the interference of immediate emotions or social expectations, which requires further validation of the authenticity of the preferences through multimodal data.

Future research can be deepened in the following areas: First, a cross-cultural comparative study can be conducted to compare the differences in users’ prioritization of design elements in different cultures, introduce a wider range of datasets including user data from different regions, cultures, and backgrounds to ensure that our analytical results have a higher degree of representativeness and reliability, and develop dynamically adjusted user categorization models to enhance generalizability. Second, machine learning technology can be combined to build user demand prediction models to capture preference changes in real time and dynamically adjust design priorities to create and optimize new product features. These limitations need to be addressed and further explored in future research so that the approach is not only applicable to female user groups but can also be extended to other neglected user groups, such as women and people with disabilities, to enhance the universality and inclusiveness of the design. In this way, the unique needs of different user groups can be more comprehensively understood and satisfied, promoting the transformation of product design from “generalization” to “personalization” and truly realizing the concept of human-centered design.