User experience mapping (UEM) is a valuable tool for analyzing user interactions with products and environments. While UEM is widely used across various design fields, accurately quantifying users' emotional needs remains a challenge. This study addresses this challenge by focusing on automotive HMIs. The researchers propose a quantitative UEM method that integrates Kansei engineering, a technique for understanding and quantifying human feelings towards products, with traditional UEM. The goal is to precisely capture and represent drivers' emotional responses to in-vehicle interface designs. This new method uses the semantic differential (SD) method to measure user emotional perceptions of the automobile interface through a questionnaire, followed by drawing user emotion curves to identify design pain points. Finally, iterative design is carried out to address these pain points. This approach enables designers to address user emotional needs more effectively, leading to more user-centric HMI design.

The literature review examines the existing HMI structures in automobiles, highlighting the growing importance of intelligent and emotional design in the automotive industry. The review covers various aspects of HMI design, including voice assistance, icon clicks, and visual perception. It also explores Kansei engineering, a methodology used to understand and quantify emotions related to products, and its application in interface design. Existing studies on Kansei Engineering show its effectiveness in quantifying user responses to interfaces. The review also discusses the use of user experience maps as tools for visualizing user emotional experiences and identifying design pain points. However, the review points out limitations in accurately expressing and quantifying emotions in existing user experience map designs, particularly within the context of smart car interfaces. This gap in research motivates the current study, which aims to develop a more precise and effective methodology for quantifying and addressing user emotional needs in automotive HMI design.

The study employed a two-phase methodology. **Phase 1: Emotional Quantification and Pain Point Identification** 1. **Sample Collection:** 18 in-vehicle HMI samples were collected, refined to 10 representative examples. 2. **Kansei Imagery Word Collection and Screening:** 196 initial Kansei imagery words (Chinese) were gathered and screened through cluster analysis and expert review, resulting in 20 final words reflecting the Geely Star HMI's design theme (“futuristic, intelligent, technological, new standard”). These 20 words were paired with their antonyms using the KJ method, which involved a focus group discussion. 3. **Questionnaire Design and Data Acquisition:** A questionnaire, employing a 7-point Likert scale based on the semantic differential method, was developed. Participants (aged 20-45, with driving experience) evaluated the 10 HMI samples using the Kansei word pairs. 161 valid questionnaires were analyzed, showing high reliability (Cronbach's α = 0.741) and validity (KMO = 0.919). 4. **User Experience Map Creation:** The mean scores from the questionnaire for each Kansei word pair at each touchpoint formed the basis of a quantitative user experience map. This map visually represented the emotional responses at different stages of HMI interaction (pre-stage observation, mid-stage interface use, and post-stage overall evaluation). **Phase 2: Iterative Design and Validation** 1. **Pain Point Analysis:** The quantitative user experience map highlighted design pain points – areas where user satisfaction was below average. These pain points focused on visual perception (e.g., inconsistent colors, lack of technological aesthetic) and functional layout (e.g., complex design, unintuitive icon placement). 2. **Iterative Design:** Based on the pain points, iterative design was conducted, incorporating user feedback and interface design principles (consistency, hierarchy, personalization). This resulted in a redesigned HMI. 3. **Validation Test:** A second questionnaire, similar to the first, was administered to a new group of participants evaluating the revised HMI. A paired t-test was conducted comparing the emotional ratings before and after redesign. The methodology incorporated quantitative methods (semantic differential, statistical analysis) with qualitative data (expert reviews, user feedback) to refine the HMI design iteratively.

The study's key findings are: 1. **Effectiveness of the Quantitative UEM:** The proposed method effectively quantified and visualized user emotional experiences related to in-vehicle HMI. The quantitative user experience map clearly showed variations and fluctuations in satisfaction levels across different touchpoints. 2. **Identification of Design Pain Points:** The initial HMI analysis revealed nine pain points with below-average emotional satisfaction scores, predominantly related to visual perception (inconsistent color schemes, lack of technological appearance, insufficient personalization) and functional layout (complexity, poor icon arrangement). The multimedia interface showed the lowest user satisfaction. 3. **Successful Iterative Design:** Iterative design, informed by the identified pain points and user feedback, significantly improved user emotional responses to the HMI. This was validated through the second questionnaire and statistical analysis (paired t-test). Specifically, the improvements focused on consistent color schemes, enhanced technological aesthetics, more intuitive icon layouts and improved personalization. Post-revision analysis showed statistically significant increases in satisfaction scores for many touchpoints, particularly for elements related to “technological” and “simple” design features. 4. **Impact of Functional Layout:** The study demonstrated that the functional layout significantly impacts users' affective experiences. A simpler, more intuitive layout reduced cognitive load and improved satisfaction. 5. **Visual Validation:** The improved user experience map showed that the emotional experience of each touchpoint improved after redesign.

The findings demonstrate the effectiveness of the proposed quantitative user experience map design method for automotive HMIs. The integration of Kansei engineering successfully quantified user emotional responses to interface elements, providing specific insights into design areas needing improvement. The iterative design process, guided by this quantitative data, resulted in significant improvements in user satisfaction. The study highlights the importance of considering emotional factors during HMI design, going beyond traditional usability measures. The visualized results of the quantitative user experience map provide designers with a clear and actionable framework for design iteration and optimization. The effectiveness of addressing the identified pain points, particularly in the areas of visual perception and functional layout, validates the method's usefulness. The findings contribute to a better understanding of user emotional experience in automotive HMIs, offering a practical and efficient design methodology for developers.

This research successfully established a quantitative user experience map design method for in-vehicle HMIs by integrating Kansei engineering and UEM. This method effectively quantifies and visualizes user emotional experiences, providing actionable insights for iterative design improvements. The study validated the method's effectiveness through a two-phase experiment, demonstrating significant improvements in user emotional satisfaction following the redesign based on the identified pain points. Future research should explore the emotional experience in dynamic driving situations and delve into deeper interface levels.

The study's limitations include the use of a stationary driving simulation environment which may not entirely reflect the complexities of real-world driving conditions. The emotional experience quantification primarily focuses on the visual aspects of the interface and further exploration of the interaction level is required. Additionally, the study mainly examined the relationship between the main interface and the first-level interfaces, needing future investigation of deeper interface levels.