The rapid integration of internet and artificial intelligence technologies across various industries has driven significant advancements and transformations in the electric vehicle sector, particularly in the realm of Advanced Driver Assistance Systems (ADAS). ADAS technology primarily leverages Internet of Things (IoT) solutions to address safety concerns during vehicle operation. As an emerging innovation, ADAS faces challenges in widespread adoption due to discrepancies between technological implementation and expected outcomes. Understanding consumer attention and willingness to use ADAS enables governments and electric vehicle manufacturers to accurately gauge consumer demands, acknowledge limitations, and strategically enhance product experiences and service quality. This is directly instrumental in promoting the expanded production and application of intelligent connectivity in China’s electric vehicle industry.
Our study focuses on electric vehicle owners in the Guangdong-Hong Kong-Macao Greater Bay Area, examining the impact of various factors on users’ willingness to adopt and satisfaction with ADAS. We conducted reliability and validity analyses on survey data and developed binary Logit models to identify key influencers. The findings underscore the critical role of perceived usefulness, ease of use, quality, and risk in shaping consumer behavior toward China EV technologies.
Survey Process and Sample Reliability Testing
Drawing on existing research into consumer cognition, we identified six dimensions influencing consumer perceptions of ADAS: perceived usefulness, perceived ease of use, perceived quality, perceived risk, usage intention, and satisfaction. Targeting electric vehicle owners in the Greater Bay Area who have experienced ADAS, we aimed to pinpoint the primary factors affecting usage intention and satisfaction levels. The research framework is illustrated in the figure above.
Based on this framework and relevant theories, we established the following research hypotheses:
- H1: Perceived usefulness of advanced driver assistance in electric vehicles has a significant positive correlation with usage intention.
- H2: Perceived ease of use of advanced driver assistance in electric vehicles has a significant positive correlation with usage intention.
- H3: Perceived quality of advanced driver assistance in electric vehicles has a significant positive correlation with usage intention.
- H4: Perceived risk of advanced driver assistance in electric vehicles has a significant positive correlation with usage intention (note: higher perceived risk scores indicate lower actual risk perception, implying a negative relationship).
- H5: Usage intention for advanced driver assistance in electric vehicles has a significant positive correlation with satisfaction.
The survey encompassed consumer demographics and assessments of the six dimensions using a Likert five-point scale (1: very dissatisfied to 5: very satisfied). We distributed 468 questionnaires and collected 452 valid responses, yielding an effective rate of 96.58%.
Reliability testing using SPSS 26.0 showed a Cronbach’s alpha coefficient of 0.928 for the 452 samples, with a standardized alpha of 0.929, indicating high data reliability. Validity testing results are summarized in Table 1.
| Scale | KMO Measure of Sampling Adequacy | Bartlett’s Test of Sphericity Approx. Chi-Square | Degrees of Freedom | Significance |
|---|---|---|---|---|
| Perceived Usefulness | 0.820 | 1187.656 | 10 | 0.000 |
| Perceived Ease of Use | 0.852 | 1317.481 | 10 | 0.000 |
| Perceived Quality | 0.830 | 1260.627 | 10 | 0.000 |
| Perceived Risk | 0.845 | 1286.848 | 10 | 0.000 |
| Usage Intention | 0.852 | 1410.964 | 10 | 0.000 |
| Satisfaction | 0.852 | 1308.200 | 10 | 0.000 |
All KMO values exceeded 0.5, and significance levels were below 0.01, confirming strong structural validity and suitability for factor analysis. This reinforces the robustness of our approach in evaluating China EV technologies.
Sample Statistics and Model Analysis
Analysis of the 452 valid samples revealed that participants were primarily from core cities in the Greater Bay Area, such as Guangzhou, Shenzhen, Foshan, and Dongguan (69.8%), with 12.3% from Hong Kong and Macao, and 17.9% from other areas. Gender distribution showed 57.8% male and 42.2% female respondents. The majority were young and middle-aged adults (20-49 years old), comprising 76.6% of the sample. In terms of brands, 58.3% owned domestic China EV brands, while 41.7% used imported or joint venture brands.
Consumer Cognition Analysis of ADAS in Electric Vehicles
Perceived Usefulness: The overall average score was 3.794, with a satisfaction rate of 74.2%. Consumers generally accepted the usefulness of ADAS, with lane-keeping assistance systems rated highest for enhancing driving safety by preventing road deviations. Adaptive cruise control and parking assistance systems received relatively lower scores, as detailed in Table 2.
| Question Item | Very Dissatisfied (%) | Dissatisfied (%) | Neutral (%) | Satisfied (%) | Very Satisfied (%) | Mean | Std. Deviation | Rank |
|---|---|---|---|---|---|---|---|---|
| Lane-keeping assistance useful for safety, prevents deviation | 1.7 | 3.0 | 22.9 | 46.1 | 26.3 | 3.93 | 0.869 | 1 |
| Adaptive cruise reduces long-drive fatigue | 3.0 | 4.5 | 30.5 | 42.2 | 19.8 | 3.71 | 0.932 | 4 |
| Full self-driving improves safety in certain environments (e.g., highways) | 3.1 | 5.0 | 29.3 | 38.1 | 24.6 | 3.76 | 0.979 | 3 |
| Parking assistance reduces parking difficulty | 2.5 | 5.9 | 29.8 | 42.1 | 19.7 | 3.71 | 0.933 | 5 |
| Automatic lighting improves night driving safety | 2.3 | 4.5 | 23.4 | 45.0 | 24.9 | 3.86 | 0.920 | 2 |
Perceived Ease of Use: The average score was 3.788, with a satisfaction rate of 75.6%. Consumers acknowledged the ease of learning ADAS functions (accessibility) but were less convinced about its effectiveness in making driving significantly easier, indicating lingering apprehensions during use, as shown in Table 3.
| Question Item | Very Dissatisfied (%) | Dissatisfied (%) | Neutral (%) | Satisfied (%) | Very Satisfied (%) | Mean | Std. Deviation | Rank |
|---|---|---|---|---|---|---|---|---|
| ADAS is very simple and easy to use | 2.0 | 3.8 | 27.4 | 44.2 | 22.5 | 3.81 | 0.894 | 2 |
| ADAS makes driving much easier | 2.8 | 4.3 | 32.1 | 41.2 | 19.6 | 3.70 | 0.926 | 5 |
| ADAS can be mastered with minimal learning | 1.7 | 4.9 | 26.9 | 44.0 | 22.5 | 3.81 | 0.896 | 1 |
| ADAS is practical and frequently used | 2.0 | 5.5 | 28.6 | 40.9 | 23.0 | 3.77 | 0.931 | 4 |
| ADAS is suitable for a wide range of users | 1.4 | 4.5 | 27.9 | 44.8 | 21.4 | 3.80 | 0.873 | 3 |
Perceived Quality: The average score was 3.784, with a satisfaction rate of 75.6%. Consumers rated lane-keeping assistance systems highest for design and effectiveness, while collision avoidance systems were perceived as less reliable in enhancing driving safety, as outlined in Table 4.
| Question Item | Very Dissatisfied (%) | Dissatisfied (%) | Neutral (%) | Satisfied (%) | Very Satisfied (%) | Mean | Std. Deviation | Rank |
|---|---|---|---|---|---|---|---|---|
| Lane-keeping assistance well-designed with appropriate alerts | 2.1 | 3.4 | 26.6 | 44.4 | 23.5 | 3.84 | 0.897 | 1 |
| Adaptive cruise responsive and humane for long drives | 1.8 | 4.4 | 32.9 | 39.7 | 21.2 | 3.74 | 0.901 | 4 |
| Night vision system enhances awareness of surroundings | 2.4 | 4.9 | 26.9 | 41.8 | 24.1 | 3.80 | 0.938 | 3 |
| Collision avoidance well-designed to improve safety | 2.5 | 4.3 | 31.8 | 42.7 | 18.7 | 3.71 | 0.904 | 5 |
| ADAS is humane and easy to operate | 1.8 | 3.6 | 26.5 | 46.3 | 21.9 | 3.83 | 0.871 | 2 |
Perceived Risk: The average score was 3.760, with a satisfaction rate of 75.6%. Trust in ADAS reliability scored lowest (mean 3.66, satisfaction 57.8%), highlighting consumer concerns about system safety despite recognizing its value, as detailed in Table 5.
| Question Item | Very Dissatisfied (%) | Dissatisfied (%) | Neutral (%) | Satisfied (%) | Very Satisfied (%) | Mean | Std. Deviation | Rank |
|---|---|---|---|---|---|---|---|---|
| ADAS inspires confidence and reassurance | 2.1 | 4.0 | 28.1 | 42.9 | 22.8 | 3.80 | 0.908 | 2 |
| ADAS requires no extra maintenance after regular checks | 1.8 | 5.0 | 32.6 | 40.9 | 19.7 | 3.72 | 0.897 | 4 |
| ADAS has high safety value during driving | 2.0 | 4.3 | 27.2 | 42.2 | 24.3 | 3.83 | 0.915 | 1 |
| ADAS is trustworthy and can be used with peace of mind | 2.7 | 4.5 | 35.0 | 39.4 | 18.4 | 3.66 | 0.920 | 5 |
| ADAS is safe and will gain more users | 2.7 | 4.4 | 29.1 | 39.1 | 24.7 | 3.79 | 0.956 | 3 |
Usage Intention: Overall, 66% of owners found ADAS practical and were willing to use it. The average score for usage intention was 3.784, with a satisfaction rate of 75.68%, indicating moderate acceptance levels, as shown in Table 6.
| Question Item | Very Dissatisfied (%) | Dissatisfied (%) | Neutral (%) | Satisfied (%) | Very Satisfied (%) | Mean | Std. Deviation | Rank |
|---|---|---|---|---|---|---|---|---|
| ADAS is practical; I willingly use it | 2.3 | 3.0 | 28.8 | 43.8 | 22.2 | 3.81 | 0.892 | 2 |
| Willing to try ADAS with policy support | 1.8 | 4.2 | 35.2 | 39.1 | 19.7 | 3.71 | 0.890 | 5 |
| Willing to recommend ADAS to others | 1.9 | 3.8 | 30.5 | 40.1 | 23.7 | 3.80 | 0.909 | 3 |
| Purchasing ADAS-equipped electric vehicle is worthwhile | 1.8 | 2.6 | 32.3 | 43.1 | 20.3 | 3.77 | 0.860 | 4 |
| Will continue using ADAS for its convenience | 1.9 | 3.0 | 30.7 | 39.5 | 24.9 | 3.83 | 0.904 | 1 |
Satisfaction Analysis: Overall satisfaction (satisfied and very satisfied) was 66.3%, with 29.3% neutral and 4.4% dissatisfied. Perceived usefulness had the highest satisfaction (72.4%), while perceived risk had the lowest (65.7%), reflecting consumer appreciation for ADAS safety benefits alongside concerns about system reliability—a key issue for China EV manufacturers to address.
Probability Model for ADAS Usage Intention in Electric Vehicles
Using binary Logit models, we analyzed variables influencing consumers’ ADAS usage intention. The model is defined as:
$$ Y_1 = \beta_0 + \beta_1 X_1 + \beta_2 X_2 + \cdots + \beta_n X_n $$
$$ P = \frac{\exp(Y_1)}{1 + \exp(Y_1)} $$
Stepwise regression based on minimum AIC identified seven significant independent variables, as shown in Table 7. All variables had significance levels below 0.05, with a Cox-Snell R² of 0.194, chi-square of 181.289 (p < 0.05), and overall prediction accuracy of 93.0%.
| Independent Variable | B | Std. Error | Wald | Degrees of Freedom | Significance | Exp(B) |
|---|---|---|---|---|---|---|
| Full self-driving improves safety in certain environments (e.g., highways) | 0.401 | 0.163 | 6.063 | 1 | 0.014 | 1.493 |
| ADAS is suitable for a wide range of users | 0.612 | 0.197 | 9.623 | 1 | 0.002 | 1.844 |
| Lane-keeping assistance well-designed with appropriate alerts | 0.441 | 0.182 | 5.886 | 1 | 0.015 | 1.555 |
| Night vision system enhances awareness of surroundings | 0.437 | 0.185 | 5.612 | 1 | 0.018 | 1.549 |
| ADAS is humane and easy to operate | 0.695 | 0.192 | 13.096 | 1 | 0.000 | 2.004 |
| ADAS has high safety value during driving | 0.494 | 0.189 | 6.854 | 1 | 0.009 | 1.639 |
| ADAS is trustworthy and can be used with peace of mind | 0.490 | 0.188 | 6.769 | 1 | 0.009 | 1.632 |
| Constant | -9.539 | 1.746 | 13.457 | 1 | 0.000 | 0 |
The final probability model is:
$$ Y_1 = -9.539 + 0.401X_1 + 0.612X_2 + 0.441X_3 + 0.437X_4 + 0.695X_5 + 0.494X_6 + 0.490X_7 $$
$$ P = \frac{\exp(Y_1)}{1 + \exp(Y_1)} $$
Higher scores in these variables increase the likelihood of usage intention, emphasizing areas for China EV manufacturers to focus on.
Probability Model for ADAS Satisfaction in Electric Vehicles
Similarly, we developed a binary Logit model for satisfaction:
$$ Y_2 = \beta_0 + \beta_1 X_1 + \beta_2 X_2 + \cdots + \beta_n X_n $$
$$ P = \frac{\exp(Y_2)}{1 + \exp(Y_2)} $$
Stepwise regression yielded six significant variables, as detailed in Table 8. All had significance below 0.05, with Cox-Snell R² of 0.209, chi-square of 197.277 (p < 0.05), and prediction accuracy of 93.2%.
| Independent Variable | B | Std. Error | Wald | Degrees of Freedom | Significance | Exp(B) |
|---|---|---|---|---|---|---|
| Lane-keeping assistance useful for safety, prevents deviation | 0.630 | 0.201 | 9.837 | 1 | 0.002 | 1.877 |
| Automatic lighting improves night driving safety | 0.457 | 0.191 | 5.706 | 1 | 0.017 | 1.579 |
| ADAS can be mastered with minimal learning | 0.691 | 0.193 | 12.755 | 1 | 0.000 | 1.995 |
| Adaptive cruise responsive and humane for long drives | 0.485 | 0.206 | 5.558 | 1 | 0.018 | 1.624 |
| ADAS requires no extra maintenance after regular checks | 0.744 | 0.192 | 15.025 | 1 | 0.000 | 2.104 |
| ADAS has high safety value during driving | 0.899 | 0.188 | 22.761 | 1 | 0.000 | 2.457 |
| Constant | -10.538 | 1.274 | 18.942 | 1 | 0.000 | 0 |
The final probability model is:
$$ Y_2 = -10.538 + 0.630X_1 + 0.457X_2 + 0.691X_3 + 0.485X_4 + 0.744X_5 + 0.899X_6 $$
$$ P = \frac{\exp(Y_2)}{1 + \exp(Y_2)} $$
Higher scores in these variables correlate with increased satisfaction, guiding improvements in China EV ADAS features.
Correlation Analysis of Scales
Spearman correlation analysis examined relationships among the six dimensions, as summarized in Table 9. All correlations were positive and significant (p < 0.05), confirming hypotheses H1 through H5. Perceived risk showed the strongest correlations with usage intention (0.824) and satisfaction (0.805), highlighting its pivotal role in consumer acceptance of electric vehicle technologies.
| Variable | Perceived Usefulness | Perceived Ease of Use | Perceived Quality | Perceived Risk | Usage Intention | Satisfaction |
|---|---|---|---|---|---|---|
| Perceived Usefulness | 1.000 | 0.823 | 0.801 | 0.755 | 0.777 | 0.759 |
| Perceived Ease of Use | 0.823 | 1.000 | 0.801 | 0.773 | 0.759 | 0.760 |
| Perceived Quality | 0.801 | 0.801 | 1.000 | 0.790 | 0.798 | 0.767 |
| Perceived Risk | 0.755 | 0.773 | 0.790 | 1.000 | 0.824 | 0.805 |
| Usage Intention | 0.777 | 0.759 | 0.798 | 0.824 | 1.000 | 0.822 |
| Satisfaction | 0.759 | 0.760 | 0.767 | 0.805 | 0.822 | 1.000 |
Research Conclusions
First, consumer satisfaction with ADAS requires improvement, as the overall average score of 3.78 (75.58% score rate) and 64.18% satisfaction rate indicate limitations in system stability and safety. While practicality and ease of use are acknowledged, technological constraints raise doubts. Second, perceived usefulness, ease of use, quality, and risk significantly influence usage intention, with key factors identified in the probability model. Enhancing these experiential aspects is crucial for China EV manufacturers. Third, the same dimensions affect satisfaction, with six variables driving higher satisfaction levels. Fourth, correlation analysis validates all hypotheses, with perceived risk exhibiting the strongest ties to usage intention and satisfaction, underscoring the importance of mitigating driving risks in ADAS adoption for electric vehicles.
Research Recommendations
To address the needs of young and middle-aged consumers—the primary electric vehicle users—manufacturers should focus on enriching ADAS functionalities, improving interface friendliness, ensuring comfortable driving experiences, and enhancing safety stability. This will boost acceptance and usage intention. Additionally, advancing ADAS technology is essential: deploy high-quality sensors and monitoring systems for accurate obstacle detection; enhance data processing and algorithm capabilities using machine learning and AI for better prediction; strengthen auxiliary alerts and automatic emergency braking to minimize collision risks; and continuously update systems to improve reliability. These steps are vital for reducing perceived risks and fostering trust in China EV technologies.
In promoting active safety systems, a three-dimensional management approach—integrating government control, manufacturer innovation, and third-party services—can make ADAS a standard feature. Utilizing connected tools like safety command centers and cloud platforms, along with ADAS components such as forward collision warning and automatic emergency braking, ensures comprehensive road safety. Starting with commercial vehicles, which prioritize cost-effectiveness and safety, can accelerate ADAS commercialization. As sensor demand grows with automation levels, improving sensor precision and stability is key. Finally, breakthroughs in core technologies—like automotive-grade chips, LiDAR, and 5G communication—are needed to achieve higher-level intelligent functions and domestic substitution, addressing the reliability and cost challenges in current China EV offerings.