The usability of diabetes self-care app is important to users, especially to the elderly and those with healthcare needs. This study conducted a usability evaluation on the diabetes self-care smartphone and found that button sizes and numeric keypads in those apps were important to correct numeric input of users. Previous studies on numeric keypad were mainly conducted on larger kiosk-sized touchscreens. Therefore, the smartphone numeric keypad design was further examined in the second and third experiments. Two keypad layouts were proposed for young adults and elderly users.

Experiment I evaluated three diabetes self-care apps through usability testing and questionnaire. Operating each app, participants performed three task scenarios, followed by a System Usability Scale (SUS) questionnaire to evaluate overall usability of the app. A high SUS score, advantages highlighted by participants via think-aloud protocol, and the user interface may serve as a reference for diabetes self-care app design to attain a higher level of usability. Following the evaluation of apps, the experiments further examined the use of the numeric keypad for data entry. The objective of the second and third experiment is to recommend different numeric keypad designs suitable for young adults and elderly users. Therefore, experiment II and III investigates objective performance and subjective user satisfaction based on the effects of smartphone numeric button design characteristics including shape, size, and spacing. Experiment II involved thirty-two healthy young adults performing a 10-digit entry task on a smartphone touchscreen to evaluate the 45 numeric keypad configurations comprised of three button shapes (circle, vertical rectangle, square), five buttons sizes with 2.5 mm increments (7.5, 10, 12.5, 15, 17.5 mm), and three button spacings (0, 1, 3 mm). Experiment II indicated that young adult user performance plateaued at the 12.5 mm size button. In addition, the 3 mm spacing error rate was significantly greater than the 0 mm spacing. In experiments II and III, the circle and square buttons resulted in significantly better performance compared to the vertical rectangle button, while the circle and square buttons did not have significant difference. The findings can be used when designing an optimal user interface for healthcare apps and smartphone numeric keypad designs.

With the rapid development of display technology, personal display terminals, such as tablet, PC and smart-phone, have become irreplaceable daily necessities in the study, work and living mode of modern people. The commonly used liquid crystal display (LCD) with Light Emitting Diode (LED) as back-light has occupied a dominant position in the field of display. While pursuing thinner, bigger and brighter, the studies on the effects of LCD light characteristics on human beings has become more and more important. Therefore, in order to more clearly understand the adaptation of users to the screen light source when using the LCD, and how to adjust the light that makes the user feel comfortable, this study analyzed the influence of brightness and color temperature changes on the comfort of users by watching LCD.

In experiment I, 12 males and 12 females were invited to observe the LCD with 18 light conditions in a completely dark environment, which were composed of 3 screen illuminance levels such as 10 lx, 100 lx, and 200 lx, as well as 6 color temperature values such as 2800 K, 3700 K, 5000 K, 6500 K, 7500 K, and 9300 K. After the experiment, the participants carried out the subjective feeling scoring of screen comfort under various light conditions. The experimental results showed that the correlated color temperature and illuminance of the screen were basically consistent with Kruithof curve. However, no matter how the illuminance changes in the middle color temperature region (6500 K- 7500 K), the comfort value of users always keeps a high level. As far as illuminance is concerned, the comfort level of 10 lx and 200 lx is lower than that of 100 lx in the absence of ambient light. Generally speaking, the influence of illuminance on comfort is not significant compared with color temperature. During the experiment, it was also found that the status attributes of operation (work & study; and leisure & entertainment) was significantly correlated with visual comfort. The maximum comfort mainly occurs in the higher color temperature range of 5000 K-6500 K in the work & study state, while the maximum comfort mainly occurs in the lower color temperature range of 3700 K-6500 K in the leisure & entertainment state. Moreover, the color temperature range of high comfort in the work & study state is relatively small, while the color temperature range of high comfort in the leisure & entertainment state is relatively loose. In addition, gender and age are significantly correlated with visual comfort. Females are more sensitive to color temperature, especially when the color temperature is low, and the female participant first shows uncomfortable reaction. The comfort of female participants appears in 6500 K-7500 K, which is slightly higher than that of male participants (5000 K-6500 K); the comfortable color temperature range of teenagers is 5000 K-6500 K, while the comfortable color temperature of participants over 60 years old is about 5000 K. Without the influence of ambient light, the elderly generally feel more comfortable with high-illuminance displays.

In experiment II, according to the results of experiment I, there is a significant correlation between the comfort of users and screen brightness, color temperature, gender, age and the nature of operation, which shows that this visual comfort is the result of a complex factor. In order to further explore the comfort situation of multiple factors, 15 males and 15 females were invited to evaluate the combination of different brightness and color temperature under three different ambient lights at random as well as the two states such as  work & study and leisure & entertainment, namely each user obtained 6 groups of experimental data through experiments. The blink frequency, pupil size and other information were recorded by using Tobii pro glasses 3 eye tracker. After correlation treatment between subjective evaluation and objective measurement, a multi-independent variable LCD display parameter prediction model based on Back Propagation Neural Network (BPNN) was established, which can more truly restore the usage of users. The trained network has a good fitting performance, and the average error is less than 10%, which can successfully and accurately predict the visual comfort of users under different parameters.

In experiment III, according to experiment II, it was found that blink frequency was significantly correlated with subjective comfort, with the strongest correlation. Therefore, the blink frequency index was used to explore the change of human visual comfort with time in experiment III, and the EOG module of Biopac MP160 Polysomnography was used to record. 12 participants were invited to calibrate the selected iPad Mini screen with a brightness meter in a dark room that can be completely protected from light. Combined with the general comfort range of experiment I, the blink frequency changes of human eyes during watching were recorded under 2 color temperature conditions (5000 K and 7500 K) and 2 illuminances (100 lx and 200 lx). The results showed that watching on a low color temperature screen for more than 1h will cause an increase in blink frequency earlier; under the condition of low brightness in the working environment, the brightness elevation of the display will effectively improve attention, but it will cause eye discomfort at the same time; in terms of the working state with tasks, fatigue occurs in advance within 1h compared with the leisure state.

In this study, the watching comfort of people for LCD screen brightness and color temperature was studied, which took into account the differences of individual users and working properties, and explored the change law of visual comfort with time. The results can provide reference for users to evaluate the comfort of watching LCD or design an adaptive light adjustment system, so as to improve the visual comfort quality of users watching LCD.

Recent years witnessed the fast development in technical advisory, Internet of things and online payment in Mainland China. Sharing economy is represented by ride-hailing services, online lodging arrangement, takeaway orderings through online platforms, sharing of medical services and resources, sharing in logistics. Those new industries and new models of the sharing economy have given impetus to restructuring of service industry, transformation of consumption and quick growth. As a wholly new green way of travel, ride- hailing constitutes a great part of online rides booking. High-quality ride hailing services guarantee the satisfaction and loyalty degree of consumers due to the online to offline (O2O) characteristics. The service quality of hitchhiling is comprised of two aspects-online and offline service quality. The gaps of service quality are formed owing to the differential awareness and expectations towards service quality, causing customers’ unsatisfactory in service. This research has been trying to find the way to improve service quality and to hoist consumer’s satisfaction. 

Since the service quality of hitchhiking is divided into the overall characteristics of online and offline, this research conducted three-stage experiments to better improve the service quality of hitchhiking and increase consumer satisfaction with the experience of hitchhiking. The first stage is to design a service quality questionnaire by apply the SERVOQUAL in PZB service quality model and the importance-satisfaction model (IS Model). The questionnaire is conducted the five major service dimensions covering online and offline service items to explore consumers regarding the current satisfaction degree of the service provided by hitchhiking and its key service quality attributes. Then the to be improved elements will come out after ranking the average difference between the importance and satisfaction until the difference test of consumers' importance and satisfaction of service quality completed. The second stage uses the Kano Model to design the content of the questionnaire and classifies consumer needs into different service quality elements to improve and enhance the quality of services. In order to better understand the two-dimensional quality attributes of the current ride-hailing services in the market and have a better definition of product quality characteristics that affect consumer satisfaction, when designing the Kano Model questionnaire content, it should first evaluate the results of the service quality questionnaire then add experience of Hitchhiking experts and the experience of frequent users as the foundation of the questionnaire as to find valuable service elements. It should be known that The Kano model can be used in conjunction with the PZB model to establish the individual importance of the quality characteristics of consumers' demand for downwind services. Therefore, in the process of product development, better priority conditions can be constructed. In the second stage of the research, Kano two-dimensional quality classification was carried out for 40 items in the five dimensions, and the consumer satisfaction coefficient index was analyzed. The increase of satisfaction coefficient and the elimination of dissatisfaction coefficient were drawn into a satisfaction coefficient index analysis matrix. At the same time, giving priority to improve the quality items in the first and fourth quadrants. Finally, the 40 quality items in the study were refined and 22 items that were prioritized for improvement were incorporated into the development of Hitchhiking 2.0 optimized version. The third stage is the expert questionnaire. The expert group is composed of 6 experts from the application software department, user experience department and computer technology department of Hangzhou Didi Travel Technology Co., Ltd. The optimized version of Didi 2.0 Demo will be conducted by the expert group who will make evaluation by using AHP evaluation scale 1, 3, 5, 7, 9 to score. After understanding consumer satisfaction with specific items of Hitchhiking service quality and brand preference, AHP can reasonably build decision Hitchhiking analysis model, systematize complex problems, construct a hierarchical structure for AHP, and build pair comparison matrices based on the results of expert questionnaires, then calculate the eigenvalues and eigenvectors of each pair comparison matrix, and verify the consistency of the matrix at the same time, Calculate the weight value of the importance of each aspect and its criteria. The research shows that among the five dimensions, "management by operators", "regulatory mechanism", and "driver service guarantee" have higher consumer demand attributes and rank among the top three. Among the specific items of the service, "Passengers can see the carpooler location on the APP", "Vehicle installation satellite positioning, detour monitoring, emergency one-click alarm and other infrastructure", "When consumers encounter danger, the surrounding car owner will display the designated point and rescue in time", "Medical emergency card for consumers and car owners to respond to emergency scenarios of protection products", "Flying car discount activities (refueling, points exchange, car purchase consultation, etc.)", "Car decoration environment Good, hitchhiking owners have high quality, good credit, and high behavioral scores. They can give priority to the order", "The hitchhiking platform provides convenient and fast payment methods", "The hitchhiking platform and car owners provide safe night travel services", "Shunfeng "Windmill City Safety Comprehensive Competitiveness Ranking" and the "increased income of star riders" are among the top ten. As a result, these needs can be set as key targets when developing new products and services for major riders. In the final research stage, identify the top-ranked Shunfeng brand in the market and verify it with the optimized version of Didi 2.0 designed and optimized in this research. Using gray relational analysis (GRA) to calculate data collected from different judges by experts can simplify complex issues and have a clearer positioning and goal for providing consumers with more quality services during the system development process. Analyze the gray relational questionnaire data filled by experts, through the standardization of the data, calculate the gray relational distance (∆0ij), calculate the gray relational coefficient (γ0ij), and finally analyze the gray relational value and AHP multi-criteria decision-making analysis of 22 items The weights of the criteria are multiplied, and the results of the research are obtained. The ranking of the pros and cons of each brand is: Didi 2.0 optimized version, Dida, Gaode, Meituan Taxi, Hello. 

The purpose of this study is to be able to apply the constructed comprehensive evaluation model to actual market development and to apply it to the evaluation of different service industries and brands under the O2O business model. The objective and standardized data of the research output can be provided as expert consulting reports to Didi Shunfeng companies. At the same time, these data can provide a reference for the most important decision analysis of enterprise designers in design and development. 

In a time of globalization, technology increasingly tends toward the ordinary, the market is easily circulated. Enterprises pay greater attention to the application of design strategy and aesthetic appearance of products, in order to strengthen their competitive advantages. With the intense competition of global market and the rapid technological advancement, the new product development becomes increasingly complicated, and relatively, designers face massive information that requires further integration. Designers need an effective mechanism to understand the influence of design strategy application and should satisfy the needs of different types of consumers using different aesthetic designs based on consumers’ lifestyles. Therefore, this study aims to use Analytic Network Process (ANP) and a two-dimensional matrix as the analysis tools, and discusses the usage situation and effect of design strategy, according to conformance quality (enterprises’ new product development performance) and design quality (the consumers’ need for new products). Furthermore, in order to understand the information regarding the advantages and disadvantages of a sense of beauty of products in the market, this study conducted a case study on air conditioners, proposed the method of aesthetic evaluation to analyze customer satisfaction, and conducted a questionnaire survey in Taiwan concerning users’ different aesthetic styles and lifestyles on air conditioners.

The research results show three main findings. First, based on the ANP results show that the order of top three items in design strategy are strengthening of market information collection and response capability (0.139), improvement of product quality level (0.135), and easy fabrication and maintenance (0.112). In addition, the ANP and the two-dimensional matrix model verified the use effect and improvement direction of design strategy. The ten design strategies can be classified into matured strategy, longitudinal improvement strategy, lateral improvement strategy, and two-way improvement strategy. The findings of this study can be provided to managers for reference in design strategy improvement. The second finding is that most people belong to the “neutral” group, and they prefer space usage, integrity and texture. The second group which has conservative and pragmatic life styles cares practical use than appearance, space usage, and size. People in third group are more active, resource-abundant and well-trained, and they have their own opinions and ideas about trends and presenting styles. They prefer the texture, space usage and integrity of the product, and they put emphasis on products’ styles, images, originality or novelty and so on. On the other hand, among the preferences, the usage of air conditioners and integrity are two most obvious factors, while geometric appearance is lowest. The third finding is related to propose the method of aesthetic evaluation. The proposed method mainly included fuzzy synthetic evaluation analysis and evaluating satisfaction and expectation value in order to calculate weight. This method could reasonably quantify the subjective and unclear information collected during the customer satisfaction survey, meet the multiple attributes of aesthetic evaluation, and shorten the survey duration in a simple and easy manner.

The results of this study are expected to propose improvement directions for design strategies. A designer can identify which design strategies can be used effectively. In addition to the evaluation methods of practicality and objectivity, this result further investigated consumers in different life styles have different preferences of aesthetic perception. Results of the study can be provided as references for product industry and designers.

Application software (app) for smartphones and tablets has become a necessity in everyday life. Up to now, millions of apps have been released worldwide; meanwhile, competition among app developers has intensified because of increasing demand. For users, an attractive interface design can get a good first impression. Besides, tens of thousands of applications have a feature in common; to be exact, they are identifiable only by means of icons. The aesthetic design of apps is greatly highlighted, and so is the relationship between the user and the interface. This study proposes the Kansei interface (KI) conceptual framework as a model to explore the abstract relationships between users and interfaces. That is, the model aims to emphasize the virtual user-interface links. Through the KI framework, the designer can clarify these abstract relationships, turning them into concrete design elements. The above-mentioned framework consists of two links (the attraction link and the motivation link) and Kansei (emotional) information. Moreover, Kansei information contains three key components: 1) enhancing the user’s positive emotions, 2) strengthening the user’s motivation for operation, and 3) achieving a balance between usability and aesthetics.

In this study, the KI framework was used to construct the research methods; interface design elements were deconstructed by the qualitative KJ method and the evaluation grid method (EGM) of Miryoku Engineering; the relationships between users and interfaces were analyzed through Quantification Type I and Type III. Finally, an experiment was designed and conducted in which the participants operated the apps. The experiment was observed and recorded, with a questionnaire survey made thereafter. Thus, how the KI framework can assist interface designers in the design process is comprehensively explored. 

The analytic results show that through deconstruction of the attraction link, the visual design of app icons can be decomposed into design styles, design categories, and design elements. Besides, the motivation link can be divided into three parts: feelings/emotions, perceptions, and design elements. Based on the above findings, the KI model of apps can be employed to interpret the relationships between users and app interfaces. This model and all the experimental data can be referenced by designers not only to discover Kansei information but also to create design values. In addition to exploring the user-interface relationships, the KI conceptual framework can be applied to explore the relationships between users and products as well as between users and services in the future.

Competing market products generally do not carry significant variations in their technological levels, but differ, rather, in their aesthetics and designs; these two factors are decisive elements in the buying decision process. Understanding the relationship between people and product aesthetic appreciation is thus a vital issue in applications regarding aesthetics and consumers. According to the principle of product shape preference, some principles are not only related to aesthetics, but also to human instinct. For instance, most people perceive a sense of threat from sharp objects, preferring curvilinear shapes instead. In other words, not only can curve preference be utilized to study human biases through aesthetics, it can also reveal people’s emotional responses to curvilinear shapes through biological reactions. Therefore, in this research, physiological data and psychological measurements were employed through experiments. The experiments studied correlations between pupillary measurements and curvature preference, all to demonstrate the principle of shape aesthetics and study the influences that altering curvature degrees may have on emotional arousal.

The research purpose was attained through two phases of experimentation: (1) To confirm whether pupil size variation can be suitably applied in product evaluations by comparing pupil response variations in observations of openers and IAPS database images (positive, neutral, or negative emotions); (2) To evaluate curvature preference in products (without altering their prototype shapes) by determining whether curvature degree and product preference exhibit an inverted-U trend and whether product size or design background can play influences on curvature preference.

The results of the experiment one indicate: Through pupil size measurement, negative and other emotional responses (neutral and positive) triggered by products can be distinguished. Because ordinary products trigger mild emotional arousal, and designer products do not trigger extremely negative emotions, pupillary measurements are thus reliable tools to conduct product evaluations. Results from the experiment two indicate: In conditions where a product’s prototype form remained unaltered, product preference and product curvature degree can indeed display an inverted-U curve, and that product size can influence how people perceive the degree of curvature. Curvature degree in larger product sizes correlates to smaller degrees of practical curvature, and ultimately influences preference. Moreover, due to design background tends to remain uninfluenced by curvature preference, especially after the peak of preference. Hence, a significant difference emerges after the peak of preference, where the preference of people with a design background declines more sharply than that of participants without a design background.

After having demonstrated that pupillary variations can suitably be applied toward product evaluation and having evaluated the correlation between curvature degree in products and preference levels (without prototype differences), the influence of altered curvature degrees on emotional arousal was lastly studied by utilizing pupillary sizes to measure variations in curvature preference for product shape. The study was facilitated through a combination of four different products and five different curvature degrees. Results indicate that variations in product form curvature do not have significant differences in arousal, regardless of pupil variations or subjective evaluations. Through subjective preference evaluation, however, one can determine the extent to which altering curvature degrees in products with varying basic shapes can carry different effects.