Keywords

1 Introduction

Environmental protection is an important survival issue that people have to face. The rapid development of economy not only brings about the improvement of people’s living quality, but also result in the greenhouse effect, serious haze and other environmental pollution problems. The number of days of serious air pollution reaches up to 60 days per year in Beijing, Tianjin and Hebei province. The environmental pollution is mainly caused by the increasing number of motor vehicles, traffic congestion and exhaust emissions. Taking Beijing as an example, according to the research results of National School of Development at Peking University in 2014, the annual losses caused by traffic congestion in Beijing is about 70 billion yuan including increased fuel costs due to congestion time, and the resulting pollution costs are also rising year by year. With the development and perfection of new energy technology represented by battery technology, more and more electric-driven vehicles appear in the lives of urban residents. As a convenient means of transportation, electric bicycle is popular with users. However, currently a large number of electric bicycles are becoming larger, heavier and faster with poor safety performance and big traffic safety risks, which result in a large number of casualty accidents. In May 2018, the Ministry of Industry and Information Technology organized the revision of the compulsory national standard of Electric bicycles–General technical requirements, which tends to solve the development problems of the electric bicycle industry in a guided way.

According to the new national standard, electric bicycles belong to non-motor vehicles, which cannot be rode in the motor lane. The maximum speed and the weight of the whole vehicle are strictly limited. The speed cannot exceed 25 km/h and the weight of the whole vehicle cannot exceed 55 kg. The motor power is 400 W and they must be equipped with pedal. According to the definition of the new national standard, electric motorcycles belong to motor vehicles, which requires driving qualification, vehicle number plate and compulsory insurance (Table 1).

Table 1. Detailed comparisons between electric bicycles and electric motorcycles
Full size table

Compared with electric motorcycles, electric bicycles have two advantages: 1. Electric motorcycles and automobiles need relatively high price of and cumbersome registration procedures, while electric bicycles are cost-effective without requirement for license and related documents, which is more preferred by consumers. 2. Electric motorcycles is dangerous due to the high speed, while electric bicycles conform the user’ requirements of safe driving due to their limited speed and weight. Therefore, as a green and convenient means of transportation, electric bicycle has become one of the irreplaceable short-distance travel modes for urban residents. Internet and industrial 4.0 era drive the transformation of product design paradigm, promote the process of manufacturing informatization, digitalization and intellectualization, which makes the design of electric bicycle evolve to intellectualization and servitization. After the data of electric bicycle is acquired by terminal, it will be analyzed and processed by the technology of cloud and big data, so as to diagnose and alarm the faults of the vehicle. The classical ones on the market are MI electric bicycle and Bywin electric bicycle, as shown in Fig. 1 below.

Fig. 1.
figure 1

Electric bicycle

Full size image

According to the changes of national standard of the electric bicycle industry, the model of the electric bicycle should be redesigned and the psychological preferences of consumers should be deeply dug. More and more consumers require the electric bicycle not only to meet the most basic functional needs, but also to meet their emotional needs. How to design an intelligent electric bicycle that meets the users’ psychological preferences has become a new issue. In order to help enterprises better understand the emotional needs of users, and feedback them to product design, so as to put forward the best product development solutions, Kansei Engineering [1] was adopted by some scholars to explore the users’ emotions in recent years, Nagamachi, a Japanese scholar, has developed the new product technology of Kansei Engineering for consumers, which is defined as a transformation technology that transforms consumers’ feelings and imaginations about products into design elements and integrates them into the development of new products. However, although KE can identify customers’ emotional attributes and associate them with design elements, it cannot provide information related to various emotional attributes on consumer satisfaction [2]. Therefore, some scholars study the emotional attributes of products with the help of KANO Model [3], which is a theory based on Herzberg’s Motivation Theory developed in 1959, also known as the Attractive Quality Theory. The influence of customers on the products and services can be understood through the Kano model questionnaire survey, which can be regarded as a reference to improve customer satisfaction [4]. However, in the process of categorizing consumers’ needs, Kano Model neglects the uncertainty of the tester’s feeling. In order to improve the accuracy of the emotions captured, this the method of fuzzy Kano model is used to explore the users, which can more accurately acquire the customers’ needs in questionnaire design so as to improve the users’ satisfaction with the products. At the same time, some scholars put forward that the attractive factors can be dug out with the help of the Evaluation Grid Method (EGM) [5], which acts as an expert evaluation method of Miryoku Engineering, aiming at understand the evaluation items and the various factors that constitute the grid through extracting consumer language. By investigating and utilizing the hierarchical structure with semantic visualization, the original items of upper and lower levels are evaluated from abstraction to representativeness [6]. In this research, with the help of the Evaluation Grid Method, the key reasons for attracting users are analyzed (mid-level), followed by the specific performance factor characteristics of the mid-level (lower level). Finally, the emotional characteristics (upper level) of the users are captured, so as to analyze the relationship between the upper perceptual image of the electric bicycle and the specific morphological characteristics of the lower level. Afterwards, the user’s quality requirements for the electric bicycle is analyzed through Fuzzy Kano, so as to complete the research on the shape design of electric bicycle.

2 Review

2.1 Miryoku Engineering

The attractive factors of products are called “miryoku” or “miryokuyinsi” in Japanese and translated into “attraction” or “attractive factors” in English. Miryoku Engineering is a technical system or preference-based design for creating attractive objects. As an expert evaluation method of Miryoku engineering, the goal of EGM is to understand various factors of evaluation items through extracting consumer language, which is also a method to understand how consumers evaluate product value through in-depth investigation. By investigating the hierarchical relationship between the image semantics of the upper level and the specific attractant features of the lower level, the original items of the upper and lower levels are evaluated from abstract to concrete, so as to visualize and immobilize the consumer value structure [6]. Therefore, the Evaluation Grid Method (EGM) is an effective method for qualitative analysis. Miryoku Engineering is used by some scholars to study the shape attraction of products. For example, Shen and Changyu [7] chose Miryoku Engineering method in order to find out the attractive factors of heavy motorcycle and used EGM (evaluation grid method) in order to find out the structure of attraction. Shen [5] aim to explore the social and cultural attractiveness of SNS game from the perspective of human emotions caused by the interaction between players and games. After interviewing knowledgeable game players with the help of the Evaluation Gird Method (EGM), it was analyzed using the I-type quantification theory. On this basis, this paper explores the relationship between the perceptual image (upper level) and the specific morphological characteristics (lower level) of electric bicycle through the Evaluation Gird Method of Miryoku engineering, and analyses the attractive factors of users to electric bicycle products, so as to explore the Attractive.

2.2 Fuzzy Kano Model

In recent years, some scholars have begun to find ways to reach human satisfaction with products. Kano et al. [8] proposed a two-dimensional quality model map, as shown in Fig. 2 below. Kano model is used to analyze user needs and develop questionnaires. The questionnaires acquire the usefulness of user needs through both positive and negative questions, that is, whether the user is satisfied if the demand is provided, whether the user is not satisfied if it is not provided. By analyzing whether there is a linear relationship between the two, the category of user needs can be identified. The results can be divided into five categories: Attractive (A), One-dimensional (O), Indifferent (I), Reverse (R) and Must-be (M), which are described as follows:

Fig. 2.
figure 2

Kano model

Full size image
  1. 1.

    Attractive Quality Element: Attractive Quality means that when products or services are available, customers will feel satisfied, but when products or services are not available, customers will also not be dissatisfied.

  2. 2.

    One-dimensional Quality Element: One-dimensional Quality means that when a product or service is available, the customer will be satisfied, but when the product or service is not available, the customer will be dissatisfied.

  3. 3.

    Must-be Quality Element: Must-be Quality means that when products or services are available, customers will take it for granted, but when products or services are not available, customers will be dissatisfied.

  4. 4.

    Indifferent Quality Element: Indifferent Quality refers to whether products or services are available or not, it will not affect customer satisfaction.

  5. 5.

    Reverse Quality Element: Reverse Quality means that when the product or service is available, the customer will not be satisfied, but when the product or service is not available, the customer will be more satisfied.

Kano model illustrates the relationship between customer satisfaction and product or service performance, as shown in Table 2 Kano’s evaluation table [9].

Table 2. Kano’s evaluation table
Full size table

However, the results obtained by traditional Kano model are not accurate, because it ignores the uncertainty caused by customers’ complex and changeable psychology, therefore, it requires the introduction of fuzzy theory to solve the fuzzy problem in reality. In 1965, L. A. Zadeh, a professor of automation from the University of California, Berkeley, first proposed Fuzzy Sets Theory. He used mathematical methods to express the uncertainty in real life [4]. In order to deal with the fuzziness of human thought, Zadeh first introduced the theory of fuzzy sets, which rationally determines the uncertainties under fuzzy or unknown conditions. The characteristics of such sets are that the distribution range of each element is between 0 and 1 [10]. Therefore, the Fuzzy Kano Model [11] has become a more valuable research subject, whose results will be more accurate. Improved and developed on the basis of traditional Kano model, Fuzzy Kano Model is introduced to capture people’s emotional factors mainly based on the consumers’ fuzzy demand for products, so as to accurately capture people’s emotions and their key quality. The relevant cases are applied in some feelings of industrial design: For example, Wang and Wang et al. [12] proposed a hybrid framework, which combines fuzzy AHP, fuzzy Kano model and zero-one integer programming (ZOIP), and integrates customer preference and customer perception into the decision-making process of product configuration. Finally, the best smart camera products in different markets were determined with the help of ZOIP by maximizing the overall customer utility (OCU) and considering the company’s pricing policy. Lee et al. [4] used the method of fuzzy questionnaires to modify Kano two-dimensional questionnaires and develop mathematical computing performance according to the quality classification of Kano two-dimensional fuzzy model. Finally, the service quality of theme park will be taken as an example. Shafia and Abdollahzadeh [13] put forward a new solution with the help of fuzzy TOPSIS and fuzzy KANO technology in their research. By prioritizing FR to improve beneficiary satisfaction, investment will be more efficient. In order to solve the uncertainties in product development process, Chen and Ko et al. [14] introduced Kano model into the fuzzy theory to improve user satisfaction. finally, the effectiveness of the method is verified in practice. On this basis, this research explores the satisfaction factors of product perceptual image through the fuzzy Kano model, so as to develop the user demand products theory that meets consumer preferences, and then enhance the shape attraction of products.

3 Method

3.1 Analysis of Shape Factors of Products

Firstly, the classical samples of electric bicycles on the market were select and in-depth interview of users were carried out through EMG method, whose research process includes the following steps: 1. Print out all the sample product pictures in A4 printing paper and display them in front of users, so that users can select the sample pictures whichever they like; 2. Ask the specific reasons why users like the sample pictures, and determine the key reason for attracting users, that is the mid-level item of attractive factors of the products; 3. Further inquiring about the specific shape characteristics of products liked by users, whose result is the lower-level characteristics of the product’s attractive factor; 4. Further inquiring about user’s psychological feelings caused by the specific attractive features of the product, and the result is the upper perceptual image of the product’s attractive factor; 5. Merge the same or similar upper (perceptual image) and lower (specific modeling features) by kJ method, and the number of times after merging is equal to the sum of times mentioned by all users before merging, and then integrate and rank by KJ method, so as to construct the evaluation grid map of the product attractive element. According to the evaluation grid map, the hierarchical relationship between the image semantics of the upper level of product attractive factors and the specific attracting features of the lower level is constructed.

3.2 Quantification Theory I

Quantification Theory is a branch of multivariate statistic put forward by Professor Lin Zhijifu of Japan. Quantification theory I mainly deals with the prediction of quantitative variables of qualitative variables. The application of this theory can quantify the problems that are difficult to study in detail and quantitatively, so as to study and discover the relations and laws between things more comprehensively [15]. Therefore, the Quantification Theory is to study the linear relationship between a set of qualitative variables X (independent variables) and a set of quantitative variables y (dependent variables), and to establish the mathematical model between them through multiple regression analysis. This research intends to use the Quantification Theory I as a tool to analyze the attractive factors of electric bicycle. Through the Quantification Theory I, the quantitative relationship between the upper perceptual image and the lower specific shape features is analyzed, and the most critical shape features affecting the upper perceptual image are explored.

According to Quantification Theory I [16], the shape design item is regarded as an item, represented by X, and the design elements in shape design item is regraded as Category, expressed as C. Suppose the M item XM has RM categories, then the xm item has cmrm categories including cm1, cm2, cm3… and suppose δi (j, k) (i = 1,……,n; j = 1,2,……,m; k = 1,2,……rj) is the value of the K category in the j item in the i sample.

3.3 Fuzzy Kano Model

Based on the fact that the traditional Kano model forces people to choose an answer, which ignores the uncertainty of human thinking, a Fuzzy Kano Questionnaire (FKQ) and a Fuzzy Kano Model (FKM) are proposed to capture users’ emotional needs. The image semantic level of attractive factors is introduced into the fuzzy Kano model to classify the quality. The user’s reactions to the situations when each image semantic expression is sufficient or inadequate are investigated through the positive and negative questionnaires, and the quality attribute of product image semantics is judged. The relationship between the attractive factors of electric bicycle and customer satisfaction is further analyzed.

The FKQ questionnaire enables the respondents to present their usual ideas and solutions more comprehensively and to match them with the human thinking model. Even if there is a little bit of feeling or thinking, the service providers will also understand through the questionnaire survey. Therefore, model and quality attribute classification of Kano are more real [4]. Hereby, the FKQ questionnaire assigns the corresponding proportion according to the user’s fuzzy emotional thinking, as shown in Table 3.

Table 3. Fuzzy Kano questionnaire
Full size table

The basic steps of the Fuzzy Kano algorithm are three steps: First, assuming that a function can include functional and dysfunctional, the matrix that can be implemented fun = [0.1 0.8 0.1 0 0], and the matrix that cannot be implemented dys = [0 0.2 0.8 0 0], and the fuzzy relation interaction matrix Z is generated by fun ⊗ dys;

$$ {\text{Z = }}\left[ {\begin{array}{*{20}c} 0 & {0.02} & {0.08} & 0 & 0 \\ 0 & {0.16} & {0.64} & 0 & 0 \\ 0 & {0.02} & {0.08} & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 \\ \end{array} } \right] $$

Secondly, after obtaining Z matrix, the value of matrix is corresponded with the attributes of requirement attribute classification table proposed by Matzler and Hinterhuber [9] according to previous literature or experience, and the membership vector T of requirement category is obtained.

$$ {\text{T = }}\left\{ {\frac{0}{\text{M}},\frac{0}{\text{O}},\frac{0.1}{\text{A}},\frac{0.9}{\text{I}},\frac{0}{\text{R}},\frac{0}{\text{Q}}} \right\} $$

Finally, the value of confidence level α is introduced for further screening. If an element in the membership vector is greater than or equal to α, the corresponding quality attribute of the element is represented by 1; otherwise, it will be represented by 0. According to the relevant scholar’s argument, a = 0.4. Therefore, the customer requirement attribute vector is transformed into T = [0, 0, 0, 1, 0, 0] through the value of a, and the quality attribute is known as indifferent quality requirement. At the same time, according to this requirement, the steps mentioned above are repeated on all users in this survey and the tendency of each customer to this requirement is counted, and the classification of customer’s demand tendency with highest frequency to this product is calculated, that is, the category of demand items corresponding to the product characteristics.

4 Discuss the Case Example

4.1 Analysis of the Attractive Factors of the Product

Firstly, the electric bicycle products that sell well on the market are collected for the research on the attractive factors of the products. 18 kinds of users’ favorite electric bicycle products were collected from the online shopping platform, magazines and related book channels as test samples. Then, the sample photos were processed through Photoshop software to make the samples into A4 size with uniform resolution format. Figure 3 below is sample example. At the beginning of the experiment, a team consisting of 20 experienced designers and users of electric bicycles needs to participate in the research of attractive factors of products, including 12 males and 8 females. The age of the research group is from 19 to 38. The original evaluation items (mid-level) and the performance characteristics (lower-level) of specific attractive factor and the affective factors (upper-level) were inquired by EGM method. The results showed that there were 23 mid-level evaluation items, 92 lower-level evaluation items and 135 upper-level evaluation items. Then similar vocabulary and specific shape features were further merged by KJ method and the results showed that there were three mid-level evaluation items, 14 lower-level evaluation items and 11 upper-level evaluation items and thus constructing the attraction level of electric bicycles. In this paper, only the design of front structure shape and saddle are discussed in the original evaluation item of the mid-level. Due to the limited research energy, the wheel hub of is not researched here. The concrete hierarchical framework of saddle and front structure shape are shown in Fig. 4.

Fig. 3.
figure 3

Sample example of electric bicycle

Full size image
Fig. 4.
figure 4

Product framework hierarchy diagram

Full size image

4.2 Fuzzy Kano Model

The relationship between product attractive morphological characteristics and product emotional imagery is found out through the above theoretical qualitative analysis. But in order to explore the customer satisfaction of product attractive factors, it is necessary to introduce the Fuzzy Kano model, classify the types of customer satisfaction according to customer demand, and define the demand into five categories: attractive demand, one-dimensional demand, indifferent demand, reverse demand and Must-be demand, so as to explore the real demand of users. In this paper, seven key evaluation items are selected from 11 perceptual images of electric bicycle according to the number of times mentioned by users. User satisfaction is discussed through Kano model and the final statistical calculation results are shown in Table 4: concise and small smart style is attractive demand, which can greatly affect user satisfaction when these two factors are outstanding. The perceptual images of dignified and stable, novelty and personality, roundness and fullness, elegance and gentleness are indifferent demands, which have no great relationship with the satisfaction of users. Technological and cool performance is one-dimensional demand, which will affect user satisfaction, but when such factors are not enough, it will reduce customer satisfaction.

Table 4. Statistical result of Fuzzy Kano
Full size table

In the design process of electric bicycle, the concise and small smart shape style need to be highlighted to improve customer satisfaction. Through the hierarchical relationship diagram of the above-mentioned evaluation grid method, the corresponding specific morphological characteristics of the style image of electric bicycle can be further qualitatively explored.

4.3 Quantification Theory

In order to accurately analyze the relationship between the specific shape and the upper-level perceptual image, the design features of the attractive factors of the electric bicycle shape are categorized to explore the contribution value of the design elements to the style image. According to Fig. 4, the shape of saddle can be divided into three design types, while the front structure shape of electric bicycle can be divided into five design types. Table 5 below is the specific shape classification table of product design.

Table 5. Classification of design elements
Full size table

Fuzzy Kano-based results of attractive demands only contains two styles of concise and small smart, so it is necessary to investigate the two image semantics of all samples with the help of SD method. Finally, the image semantics scores of all samples are obtained, and then the results of the questionnaire are sorted out, and the average number of image vocabulary is calculated. Afterwards, the multivariate regression analysis is carried out with the perceptual image vocabulary as dependent variables and product modeling feature categories as independent variables, so as to obtain the corresponding category score of type elements. Each category score is the contribution value of each sample type element to perceptual image. Then statistical analysis of Partial Correlation coefficient is carried out to obtain the score of Partial Correlation coefficient. Accordingly, the product perceptual evaluation matrix is shown in Table 6, and the final numerical results of the relationship between product morphological characteristics and perceptual intention are shown in Tables 7 and 8.

Table 6. Perceptual evaluation matrix
Full size table
Table 7. Analysis results of the relationship between design elements and perceptual vocabulary (concise)
Full size table
Table 8. Analysis results of the relationship between design elements and perceptual vocabulary (small and smart)
Full size table

In Table 7, it can be seen that the front structure shape has greater effect on the design items that affect the perceptual image of concise design style, and the partial correlation coefficient score is 0.863. In the specific shape design of the front structure of electric bicycle, S24 and S21 are positive values, among which the bionic design of S24 reaches 0.367, which is most inclined to the concise style. In the saddle design of electric bicycle, the arc shape is 0.334, which can most affect the concise style. At the same time, according to Table 8, in the design item of small and smart style image, the partial correlation coefficient of saddle is 0.852, among which the X12 curved saddle can most affect the perceptual style characteristics of small and smart products, while the bionic design of the front structure can be up to 0.551. Accordingly, a bionic style design of the front structure can trigger small and smart style image.

5 Conclusion

With electric bicycle as an application case, this research extracts its shape attractive factors with the evaluation grid method, and merges similar items into three mid-level evaluation items, 14 lower-level evaluation items and 11 upper-level evaluation items, thus forming the hierarchical relationship between the shape and emotional image of electric bicycle. Then the quantitative relationship between perceptual image and specific design features is explored through the quantitative theory I, and then by introducing it into the fuzzy Kano model, the attractive demand is analyzed. This research method can aid product developers in accurately analyzing consumers’ perceptual demands, so as to actively guide designers for innovative design of electric bicycles.