Content uploaded by Guanghong Xie
Author content
All content in this area was uploaded by Guanghong Xie on Nov 05, 2024
Content may be subject to copyright.
Optimization of Interactive Intelligent Metro Public
Facilities: Design Strategies and Satisfaction
Analysis Based on User Experience Theory
Guanghong Xie
College of Fashion and Design
Donghua University
Shanghai, China
2221931@mail.dhu.edu.cn
Xiyuan Wang*
College of Fashion and Design
Donghua University
Shanghai, China
*shbjy001@163.com
Abstract—With the promotion of China’s intelligent
transportation policy, the impact of interactive intelligent
systems and technologies in metro public facilities on
passengers’ travel satisfaction has been increasingly
emphasized; however, the current mechanism of influence
between the level of intelligence and user satisfaction of metro
public facilities has not yet been deeply explored. Therefore,
this study takes Shanghai metro transportation as the
background, and based on the extended technology acceptance
model, seven research hypotheses based on the satisfaction of
interactive intelligent metro public facilities are constructed by
introducing four variables, namely, functional experience,
content experience, interactive experience, and emotional
experience, in user experience theory. Secondly, the study
analyzed the data from the 299 valid samples collected through
reliability, validity, normality, correlation, heat map, model
fitness and structural equation modeling. The results show that
all the research hypotheses are valid and there is a progressive
relationship between different experience factors, in which
functional experience affects perceived ease of use, emotional
experience affects perceived usefulness, and perceived
usefulness directly affects satisfaction. Based on the above
research, four principle design strategies are innovatively
proposed: “de-complicating and simplifying,” “approachable,”
“striving for excellence,” and “caring.” Based on these
strategies, a set of interactive smart metro public facilities and
supporting HMIs and a set of WeChat mini-program HMIs are
designed. Finally, the effectiveness of the four-principle design
strategy is verified by conducting a usability assessment of this
design solution.
Keywords—interactive intelligent system, technology
acceptance model, structural equation modeling, data analysis,
human-computer interaction interface
I. INTRODUCTION
Within the framework of the 14th Five-Year Plan,
intelligent transportation stands out as a crucial focus,
particularly in the domain of interactive intelligent systems
and technologies. Notably, metro public facilities serve as
the linchpin infrastructure in urban transportation systems,
with their dynamic application of interactive intelligent
metro public facilities (IIMPF). IIMPF, an avant-garde urban
transportation infrastructure, leverages cutting-edge
technology to elevate the user experience (UX), efficiency,
and convenience of metro systems through sophisticated
design and information interaction. Research reveals that
several first-tier Chinese cities have successfully
implemented IIMPF, yielding substantial benefits. For
instance, Shenzhen Metro Line 20 has deployed intelligent
systems across various domains, including trains, signals,
power supply, communications, ticketing, passenger flow
monitoring, operation scheduling, and equipment
maintenance. This implementation achieves fully automatic
driving, unmanned ticketing, unmanned checking, and
unattended operations, resulting in a 10% boost in
operational efficiency and a 15% reduction in energy
consumption costs. Similarly, the second phase of Tianjin
Metro Line 6 has incorporated an intelligent station
integrated operation platform, environmental control system,
and energy consumption measurement system, leading to
intelligent station management and energy-efficient
operations, resulting in a more than 20% reduction in energy
costs. Guangzhou Metro Line 18 adopts intelligent features
like face-swipe station entry, passenger flow heat map
observation, one-key station switching, and intelligent
lighting, fostering passengers’ intelligent travel experiences
and services and contributing to a remarkable 30% increase
in passenger satisfaction. Shanghai’s commitment to
constructing high-quality integrated transportation aligns
with its vision of a cosmopolitan city that is ‘people-oriented,
efficient, intelligent, green, and resilient.’ The Shanghai
Municipal Government, as outlined in a White Paper on
Transportation Development, aims to enhance metro
passenger satisfaction from 85.5 points in 2020 to a target of
87 points by 2023. In pursuit of this goal, Shanghai has
implemented a series of measures in the interactive
intelligent systems and technology of metro public facilities.
In academia, Yang et al. [1] emphasized the fundamental
nature of rest facilities in metro stations and ongoing
improvement efforts. Yao et al. [2] observed deficiencies in
interchange metro stations in some developing countries,
impacting service performance and operational efficiency.
Zhang [3] evaluated urban rail transit service quality through
passenger satisfaction. While existing research on metro
public facilities has yielded rich results, a significant gap
remains in understanding the integration of interactive
intelligent systems and technologies. Current research
primarily focuses on facility usability, operation, and service,
with insufficient attention to the in-depth impact mechanism
of interactive intelligent systems on passenger satisfaction in
metro public facilities. To establish a mature IIMPF
satisfaction factor model, there is a need for a profound
exploration of passenger satisfaction and design strategies
based on UX theory. A search reveals a limited literature on
IIMPF satisfaction using structural equation modeling, with
few studies employing UX as a latent variable to develop
IIMPF-influencing factors. Therefore, existing studies may
not adequately explain and discuss this specific situation.
Diverging from other research, this study uses Shanghai’s
interactive intelligent metro public facilities as a concrete
example, analyzing passenger satisfaction based on the TAM
and integrating UX theory innovatively. Unlike past studies
that emphasize operational efficiency and service quality,
1
2024 International Conference on Interactive Intelligent Systems and Techniques (IIST)
979-8-3503-7442-1/24/$31.00 ©2024 IEEE
DOI 10.1109/IIST62526.2024.00022
this study underscores passenger satisfaction driven by
interactive intelligent systems and technologies under the
human-centered design concept. The incorporation of the UX
variable allows for a better observation of the impacts of
individual behavioral characteristics when using IIMPFs.
The above analysis showcases the dynamic and evolving
nature of metro public facilities within interactive intelligent
systems. These systems and technologies have
revolutionized passengers’ travel experiences and have far-
reaching social implications. Consequently, this paper
addresses the following research questions:
RQ1: Key factors influencing IIMPF through UX theory.
RQ2: Impact of interactive intelligent systems on metro
facilities and their effect on passenger satisfaction (SA).
RQ3: Evaluation of the Technology Acceptance Model
(TAM) with Function Experience (FE), Content Experience
(CE), Interaction Experience (IE), and Emotion Experience
(EE) factors on SA in IIMPF usage, manifestations, and
derived design strategies.
II. OVERVIEW AND RESEARCH HYPOTHESES
A. Interactive Intelligent Metro Public Facilities (IIMPF)
Globally, IIMPF, encompassing innovations like smart
navigation, digital ticketing, security, interactive kiosks, seat
awareness, and voice assistants, is gaining rapid popularity.
Academic focus spans UX, technology adoption, data
privacy, and social impact, with IIMPF research witnessing a
growing trend despite being in its early stages.
B. User Experience Theory (UX)
In 1995, Norman et al. [4] proposed and popularized the
concept of UX and extended its application to all aspects of
products and services. According to Luo [5], UX is the
psychological feeling that users have during the process of
using a product (including material and non-material
products) or a service. Han argued [6] that UX has multiple
characteristics that can be embedded in a scenario, but a
structured approach to integrating these characteristics is
lacking. Wang et al. [7] proposed to categorize UX into four
dimensions (FE, CE, IE, and EE), and Liu et al. [8]
investigated the factors influencing user stickiness of online
visual arts platforms based on the four dimensions of UX.
Wang et al. [7] have identified that FE in the UX contains
information access, data logging, program performance,
privacy protection, and copyright protection; CE contains
content type, content usability, and overall content quality;
and IE contains interface layout, interface aesthetics, picture
clarity, and mode of operation by carrying out an evaluation
system. The EE experience contains user feedback, user
rewards, personalized needs, and use of services.
C. Technology Acceptance Model (TAM)
In 1989, Davis introduced the TAM [9], a key framework
explaining the acceptance and utilization of new technologies.
It incorporates perceived usefulness (PU) and perceived ease
of use (PEU) as user attitudes, predicting their acceptance of
the technology.
D. Research Hypotheses
In the perspective of this study, FE refers to the basic
functional aspects perceived by users when using IIMPF; CE
refers to the user’s feelings about the content provided by
IIMPF; IE refers to the interaction process between the user
and IIMPF; and EE refers to the emotional feelings
generated by users when using IIMPF. Based on the above
analysis, this study hypothesizes that when users are using
IIMPF, it should be started by FE, deepened by CE, touched
by IE, and ended by EE, thus completing the experience
closed loop. According to relevant information systems and
technology studies [13,14,15], it has been found that the
degree of users' perception of functionality will directly
affect the degree of favorability of users' perception of
content. In addition, comfortable content delivery can lead to
a smooth interaction experience. Finally, smooth and
unobstructed interaction makes users happy. Therefore, the
following research hypotheses are proposed:
H1: FE significantly influences CE.
H2: CE significantly influences IE.
H3: IE significantly influences EE.
Liu et al. [8] have confirmed the importance of the four
dimensions of FE, CE, IE, and EE in improving service
quality. Among them, FE is the starting point of the usage
process, which directly affects users’ judgment on the
efficiency and operational difficulty of IIMPF; EE is the
endpoint, which affects users’ perception of the demand
value of IIMPF. Studies have shown that technological
systems that elicit positive emotions from users are often
found to have greater practical value [16]. The need for
functional design to be easy to use is consistently
emphasized [17]. Therefore, the following research
hypotheses are proposed:
H4: EE significantly influences PU.
H5: FE significantly influences PEU.
Research has shown [9] that PU is the extent to which the
use of a technology or product improves productivity; PEU
is the ease with which a user can use a technology or product.
In a study of middle school students' acceptance of e-
learning technology, it was found that PEU significantly
influenced PU, while PU significantly influenced SA [18]. In
summary, in the perspective of this study, PU refers to users’
perceptions and expectations of the actual benefits when
experiencing IIMPF; PEU refers to the ease or difficulty of
using IIMPF; and SA refers to the comprehensive evaluation
of users’ experience with IIMPF. Therefore, the following
research hypotheses are proposed:
H6: PEU significantly influences PU.
H7: PU significantly influences SA.
III. RESEARCH METHODOLOGY
A. Development of Test Scales
This study enhances the maturity scales using both
qualitative and quantitative methods. Initially, three experts
were engaged to review and refine the questionnaire,
addressing any ambiguities. Each item was assessed on a 7-
point Likert scale (1=strongly disagree, 7=strongly agree).
References for the scales, including FE, CE, IE, and EE from
Wang et al. [7], Laugwitz et al. [10], Liu et al. [8], PU, and
PEU from Chiu et al. [11], and SA from Kim et al. [12].
B. Data Collection and Sample Characterization
The questionnaire was distributed through electronic
channels on “QQ,” “WeChat,” and “Xiaohongshu” platforms,
yielding 330 responses. After screening (removing samples
2
with excessively short filling times and overly consistent
data), 299 valid samples were obtained, resulting in an
effective recovery rate of 91%. As per Table I, the tested user
group predominantly comprises individuals aged 18–30, with
46.5% holding a bachelor’s degree.
TABLE I. DISTRIBUTION OF SAMPLE CHARAC TERISTICS (SOURCE:
DRAWN BY THE AUTHOR)
Variables O
p
tion Fre
q
uenc
y
Percenta
g
e
Gender Male 129 43.1
Female 170 56.9
Age
<18 35 11.7
18-30 186 62.2
31-44 76 25.4
>44 2 0.7
Academic
qualifications
Below Bachelor’s
Degree 103 34.4
Bachelor’s Degree 139 46.5
Master’s Degree 48 16.1
Doctoral Degree 9 3
IV. DATA ANALYSIS AND RESULTS
This study employed SPSS for reliability tests,
descriptive statistics, normality tests, and correlation
analyses. Correlation heat maps were generated on the Tutu
Cloud Analytics website. AMOS software was utilized for
model fitness tests, convergent validity tests, combinatorial
reliability tests, and discriminant validity tests. Structural
equation modeling (CB-SEM) analyses were also conducted.
A. Confidence Validity Test
1) Reliability test: According to Table II, the Cronbach
alpha coefficients of each variable are greater than 0.7, and
the reliability is more credible.
TABLE II. RELIABILITY TEST (SOURCE: DRAWN BY THE AUTHOR)
Variable Cronbach Al
p
ha Number of terms
FE 0.802 4
FE 0.845 4
IE 0.778 3
EE 0.77 3
PU 0.781 3
PEU 0.794 3
SA 0.815 4
AGGREGATE 0.896 27
2) Model fitness test: According to Table III, CMIN/DF
is 1.684, which is excellent; RMSEA is 0.048, which is
excellent; and the values of IFI, TLI, and CFI are all greater
than 0.9, which is excellent. Therefore, the model has good
fitness.
TABLE III. MODEL FITNESS TEST (SOURCE: DRAWN BY THE AUTHOR)
Indicator Reference standard
Measuremen
t results
CMIN/DF Greater than 1 less than 3 is excellent, and
greater than 3 less than 5 is good 1.684
RMSEA Less than 0.05 is excellent, and less than
0.08 is good 0.048
IFI Greater than 0.9 is excellent, and greater than
0.8 is good 0.948
TLI Greater than 0.9 is excellent, and greater than
0.8 is good 0.937
CFI Greater than 0.9 is excellent, and greater than
0.8 is good 0.947
3) Convergent validity and combined reliability test:
Table IV shows AVE values exceeding 0.5 and CR values
exceeding 0.7 for each dimension, indicating strong
convergent validity and reliability. In Fig. 1, we show the
validated factor analysis CFA model plot.
TABLE IV. CONVERGENT VALIDITY TEST AND COMBINED
RELIABILITY TEST OF EACH DIMENSION (SOURCE: DRAWN BY THE AUTHOR)
Path coefficient Estimate AVE CR
FE1 <--- FE 0.624
0.51 0.805
FE2 <--- FE 0.765
FE3 <--- FE 0.722
FE4 <--- FE 0.737
CE1 <--- CE 0.644
0.583 0.847
CE2 <--- CE 0.762
CE3 <--- CE 0.819
CE4 <--- CE 0.816
EE1 <--- EE 0.792
0.529 0.77
EE2 <--- EE 0.723
EE3 <--- EE 0.661
PU1 <--- PU 0.771
0.545 0.782
PU2 <--- PU 0.731
PU3 <--- PU 0.712
PEU1 <--- PEU 0.772
0.563 0.794
PEU2 <--- PEU 0.757
PEU3 <--- PEU 0.721
SA1 <--- SA 0.753
0.527 0.816
SA2 <--- SA 0.651
SA3 <--- SA 0.724
SA4 <--- SA 0.771
IE1 <--- IE 0.662
0.542 0.779
IE2 <--- IE 0.765
IE3 <--- IE 0.776
4) Tests of discriminant validity: In Table V, the bolded
diagonal represents the square root of the AVE. The
standardized correlation coefficients between each pair of
dimensions are lower than the corresponding square root of
the AVE, indicating satisfactory discriminant validity
between dimensions.
TABLE V. TESTS OF DISTINCTIVE VALIDITY (SOURCE: DRAWN BY
THE AUTHOR)
Variable FE CE IE EE PU PEU PEU
FE 0.714
CE 0.632
0.764
IE 0.29 0.531
0.736
EE 0.448 0.641 0.387 0.727
PU 0.459 0.299 0.138 0.268 0.738
PEU 0.506 0.369 0.246 0.195 0.666 0.750
SA 0.527 0.648 0.305 0.722 0.558 0.428 0.726
3
Fig. 1. Validated factor analysis CFA model plot (Source: drawn by the
author).
B. Descriptive Statistics and Normality Test
In Table VI, variable mean scores range from 4.68 to
5.97, indicating an above-average understanding of IIMPF,
scored on a 1 7 positive scale. The skewness (absolute
value İ 3) and kurtosis (absolute value İ 8) coefficients
meet the criteria for a close approximation to a normal
distribution, as shown in Table VI.
TABLE VI. NORMALITY TEST TABLE (SOURCE: DRAWN BY THE AUTHOR)
Dimension Variable Mean Standard deviation Skewness Kurtosis
FE
FE1 5.19 1.097 -0.775 1.127
FE2 5.23 1.116 -0.495 0.201
FE3 5.31 1.141 -0.537 0.1
FE4 5.42 1.112 -0.779 1.084
CE
CE1 4.95 1.157 -0.588 0.152
CE2 5.13 1.198 -0.418 0.032
CE3 5.05 1.228 -0.506 0.102
CE4 5.11 1.199 -0.644 0.586
IE
IE1 4.73 1.134 -0.056 -0.176
IE2 5 1.154 -0.231 -0.201
IE3 4.82 1.143 -0.299 0.029
EE
EE1 4.93 1.199 -0.528 0.362
EE2 4.68 1.425 -0.331 -0.249
EE3 5.13 1.211 -0.52 0.308
PU
PU1 5.74 0.94 -1.004 2.585
PU2 5.82 0.984 -0.632 0.236
PU3 5.82 0.993 -0.716 0.71
PEU
PEU1 5.97 0.951 -0.804 0.435
PEU2 5.86 0.995 -0.82 1.024
PEU3 5.77 1.003 -0.823 1.029
SA
SA1 5.33 1.12 -0.688 0.669
SA2 4.86 1.242 -0.225 -0.357
SA3 4.9 1.259 -0.405 0.04
SA4 5.41 1.062 -0.713 0.895
C. Correlation Analysis and Heat Map
In Table VII, all correlation coefficients (r) between
variables are positive, indicating significant positive
correlations. Fig. 2 illustrates significant positive correlations
in orange and purple colors, with darker colors denoting
stronger correlations.
4
TABLE VII. PEARSON CORRELATION ANALYSIS BETWEEN THE DIMENSIONS (SOURCE: DRAWN BY THE AUTHOR)
Dimension FE CE IE EE PU PEU PEU
FE 1
CE 0.550** 1
IE 0.243** 0.420** 1
EE 0.364** 0.523** 0.291** 1
PU 0.363** 0.253** 0.105 0.207** 1
PEU 0.395** 0.309** 0.186** 0.158** 0.522** 1
SA 0.435** 0.530** 0.241** 0.565** 0.430** 0.337** 1
Fig. 2. Correlation heat map analysis (Source: drawn by the author)
D. Analysis of Structural Equation Modeling
Before hypothesis testing, a model fitness test was
conducted to test the relationship between the variables in the
structural equation model. The results of the test were
CMIN/DF = 2.367, RMSEA = 0.068, IFI = 0.889, TLI =
0.875, and CFI = 0.888. See Table IX for a good model fit.
As reported in Table VIII and Fig. 3, hypotheses H1, H2, H3,
H5, H6, and H7 are all valid at the 0.001 level.
TABLE VIII. STRUCTURAL EQUATION MODELING ANALYSIS (SOURCE: DRAWN BY THE AUTHOR)
Assum
p
tions Relationshi
p
Path coefficient S.E. C.R. P Result
H1 CE <--- FE 0.634 0.098 7.14 *** Supported
H2 IE <--- CE 0.579 0.085 6.828 *** Supported
H3 EE <--- IE 0.458 0.091 5.871 *** Supported
H4 PU <--- EE 0.249 0.049 3.971 *** Supported
H5 PEU <--- FE 0.536 0.087 6.661 *** Supported
H6 PU <--- PEU 0.661 0.073 8.491 *** Supported
H7 SA <--- PU 0.624 0.098 8.017 *** Supported
5
Fig. 3. Plot of path coefficients for structural equation modeling (Source: drawn by the author)
V. DISCUSSION
A. Design Strategy
This study aims to explore the influence of UX on IIMPF
user behavior and satisfaction, utilizing the TAM [9].
Through data analysis, a cascading relationship was
identified among UX factors: FE, CE, IE, and EE [7].
Building on this, four innovative principles were proposed.
1) Principle of De-complicated and Simple: The study
establishes FE’VVLJQLILFDQWLPSDFW RQ &( + ȕ
indicating that users’ engagement with IIMPF,
encompassing factors like information access, data logging,
program performance, privacy, and copyright protection,
directly influences content assessment. Additionally, FE
VLJQLILFDQWO\LQIOXHQFHV38 + ȕ UHYHDOLQJ WKDW
ease of information access and service robustness impact
users’ evaluations of IIMPF usability. PEU significantly
LPSDFWV38+ȕ KLJKOLJKWLQJWKDW,,03)VHUYLFH
ease profoundly influences evaluations of metro facility
services. The study identifies functional issues, suggesting
enhancements through interactive services like AI robots or
voice assistants to ensure information access reliability.
Effective communication with facility operators is crucial
for backend support, emphasizing FE’s role in the critical
path of PEU. Designers should prioritize simplicity for easy
function access [17], with PEU holding the most significant
path coefficient for user PU assessment. This understanding
guides designers in optimizing IIMPF for FE, enhancing
interface friendliness, simplifying operations, and improving
overall user SA.
2) Principle of Approachable: The study establishes
CE’V VLJQLILFDQWLPSDFW RQ ,( + ȕ LQGLFDWLQJ
that content accessibility and quality during IIMPF use
directly influence interface element evaluations. Issues like
users struggling to locate exit gates reveal limitations in the
navigation system. Balancing commercial interests and
information delivery is crucial for managers, suggesting
optimization of billboards and app layouts by designers to
enhance the UX [14].
3) Principle of Excellent: The study confirms the
VLJQLILFDQWLPSDFWRI ,( RQ ((+ ȕ LQGLFDWLQJ
that optimizing the interface in IIMPF services directly
influences user feedback, returns, personalized needs, and
usage evaluation. Technology advancement has matured
system interaction modes, and in IIMPF, information
reminder devices are continuously optimized using
technologies like virtual reality and artificial intelligence to
enhance IE and user emotion. Designers should adopt a
human-centered approach and leverage cutting-edge
technologies for service improvement. In payment and
arrival reminder interfaces, a focus on mobile-assisted
program layout, aesthetics, clarity, and operation can
enhance the mobile UX, increasing user SA of IIMPF [6].
4) Principle of Caring: The study concludes that EE
VLJQLILFDQWO\ LPSDFWV 38 + ȕ LQGLFDWLQJ WKDW
user feedback, individualized needs met during IIMPF use,
service incentives, and the level of care directly influence
how users evaluate IIMPF in terms of enhancing their
experience and achieving their goals. Furthermore, PU
VLJQLILFDQWO\DIIHFWV 6$ + ȕ HPSKDVL]LQJWKDW
the effectiveness of IIMPF services directly influences
users’ overall evaluation [5], particularly in enhancing travel
experience and efficiency. Injecting cultural interest into
public facility design has been found to alleviate the
monotony of metro waiting times, enhancing the overall SA
of metro public facilities. The successful integration of
regional culture, storytelling, and animation design at the
Shanghai Disney metro station serves as an example.
Designers should thoroughly explore the regional culture
and skillfully incorporate it into the design, a crucial
strategy for enhancing the UX. Based on the results in Table
1, it is reflected that the demographic characteristics of the
current sample data are mainly focused on commuters aged
18–30 with a bachelor's degree, and therefore the above
strategies lack consideration of diversity. We suggest that
6
designers should consider vulnerable groups such as the
elderly and visually impaired, take into account their
emotional experience, and incorporate their feedback to
ensure a more inclusive design of IIMPF services.
B. Design Practice
Based on the principles of De-complicated and simple,
Approachable, Excellent, Caring, a closed loop of experience
design was constructed. The design includes the exterior
design of the IIMPF, the accompanying HMI, and a set of
HMIs for the WeChat app, see Fig. 4.
Fig. 4. Interactive Intelligent Metro Public Facility design scheme (Source:
drawn by the author)
C. Usability Assessment
To validate the effectiveness of the design solution
guided by the design strategy in enhancing the metro travel
experience, an online survey was conducted. The
questionnaire comprised two parts: the first presents the
complete design scheme with renderings and descriptions,
and the second contains the scale. After excluding samples
that did not meet the criteria, 150 valid responses were
obtained (Table IX). Scores for each dimension ranged from
4.86 to 5.53, indicating respondents were moderately to
highly satisfied with the IIMPF service design solution.
Usability analysis revealed that the design strategies
effectively improved IIMPF SA.
TABLE IX. ASSESSMENT ANALYSIS OF UX SATISFACTION SCALE
(SOURCE: DRAWN BY THE AUTHOR)
Assessment content No. Mean value Standa rd deviation
FE
FE1 5.27 1.197
FE2 5.35 1.093
FE3 5.43 1.172
FE4 5.49 1.122
CE
CE1 4.96 1.253
CE2 5.12 1.226
CE3 5.02 1.201
CE4 5.15 1.157
IE
IE1 5.47 1.021
IE2 4.99 1.19
IE3 5.09 1.181
EE
EE1 4.97 1.176
EE2 4.86 1.366
EE3 5.15 1.278
SA
SA1 5.39 1.135
SA2 4.97 1.193
SA3 5.03 1.201
SA4 5.53 0.953
VI. CONCLUSION
This study employs SEM to deeply investigate the
influence mechanism of SA on IIMPF, focusing on the user
experience of IIMPF in Shanghai and aiming to provide
targeted design strategies for the actual design of IIMPF. In
terms of theory, this study constructs a model of SA
influencing factors based on four UX dimensions and
proposes four design principles based on experimental results.
The study provides some theoretical application value for
related scholars studying IIMPF and intelligent
transportation. On the practical side, the proposed strategy
consisting of the four design principles provides a reference
value for metro companies, hardware suppliers, software
developers, and urban planning stakeholders to carry out
related work. First, Metro Transit should make use of good
data analysis and user feedback mechanisms to continuously
optimize the UX of IIMPF from a diversified perspective.
Second, hardware vendors should prioritize good robustness,
sustainability, and energy efficiency based on AI technology
scenarios to ensure good environmental impacts and stable
operating costs. Third, software developers should
incorporate recommender systems and machine learning
algorithms to optimize information and services and
safeguard user privacy. Fourth, urban planners should
consider the seamless integration of IIMPF with the urban
landscape. On the social level, this study is in line with the
14th Five-Year Plan. Finally, this study has some limitations,
for which we suggest that subsequent scholars conduct
additional research. First, this study used a web-based
questionnaire to collect sample data, which is not strongly
scientific about users' real intentions. We suggest that future
scholars adopt one-on-one, in-depth interviews or collect
objective user behavioral data through experimental
monitoring equipment. Second, the sample size of this study
is too small, which limits further in-depth discussion. Even
though the sample size of this study meets the required
sample size for SEM analysis, a larger sample size should be
required for measurement due to the special nature of this
research topic. In this regard, we suggest that subsequent
scholars should increase the size of the sample data
collection. Third, this study used the TAM model in
7
combination with the UX four-dimensional model to
construct a model that focuses on explaining the
psychological satisfaction of commuters under each
dimension of experience when using the IIMPF technology.
We suggest that future scholars incorporate more variables to
discuss the multivariate nature of IIMPF, for example, trust
and privacy.
REFERENCES
[1] T. Yang and B. Shao, “Design of Subway Station Rest Facilities
Based on TRIZ Theory,” in Proceedings of the 2nd International
Conference on Information, Control and Automation, ICICA 2022,
December 2-4, 2022, Chongqing, China, 2023.
[2] L. Yao, L. Sun, W. Wang, and H. Xiong, “Adaptability analysis of
service facilities in transfer subway stations,” Mathematical Problems
in Engineering, vol. 2012, pp. 1–12, Jan. 2012.
[3] H. Zhang, “Evaluation of urban rail transit service quality based on
passenger satisfaction,” Master’s Thesis (Beijing Jiaotong University).
[4] D. Norman, J. Miller, and A. Henderson, “What you see, some of
what’s in the future, and how we go about doing it: HI at Apple
Computer,” in Conference companion on Human factors in
computing systems, 1995, p. 155.
[5] S. Luo, R. Gong, and S. Zhu, “User experience oriented software
interface design of handheld mobile devices,” Jisuanji Fuzhu Sheji Yu
Tuxingxue Xuebao, vol. 22, no. 6, pp. 1033–1041, Jul. 2010.
[6] H. Ting and K. Sato, “User experience and behaviour research based
on design information framework,” Journal of Northwest University
(Natural Science Edition), vol. 12, no. 20, pp. 9679-9686, 2012.
[7] Y. Wang and Y. Liu, “Evaluation System of CG Art Communication
Platform Based on User Experience,” IEEE Access, vol. 10, pp.
128742-128753, 2022.
[8] Y. Liu and Y. Wang, “Empirical study on the factors affecting user
stickiness of online visual art platform from the perspective of user
experience,” IEEE Access, vol. 11, pp. 60763–60776, Jan. 2023.
[9] F. D. Davis, “Perceived usefulness, perceived ease of use, and user
acceptance of information technology,” Management Information
Systems Quarterly, vol. 13, no. 3, p. 319, Sep. 1989.
[10] B. Laugwitz, T. Held, and M. Schrepp, “Construction and evaluation
of a user experience questionnaire,” in Lecture Notes in Computer
Science, 2008, pp. 63–76.
[11] W. Chiu and H. Cho, “The role of technology readiness in
individuals’ intention to use health and fitness applications: a
comparison between users and non-users,” Asia Pacific Journal of
Marketing and Logistics, vol. 33, no. 3, pp. 807–825, Jul. 2020.
[12] M. Kim and H. Qu, “Travelers’ behavioral intention toward hotel
self-service kiosks usage,” International Journal of Contemporary
Hospitality Management, vol. 26, no. 2, pp. 225–245, Feb. 2014.
[13] F. D. Davis, “User acceptance of information technology: system
characteristics, user perceptions and behavioral impacts,”
International Journal of Man-Machine Studies, vol. 38, no. 3, pp.
475–487, Mar. 1993, doi: https://doi.org/10.1006/imms.1993.1022.
[14] P. E. Paredes et al., “Driving with the Fishes: Towards Calming and
Mindful Virtual Reality Experiences for the Car,” Proc. ACM Interact.
Mob. Wearable Ubiquitous Technol., vol. 2, no. 4, pp. 1–21, Dec.
2018, doi: 10.1145/3287062.
[15] Gonçalves, V. P., Giancristofaro, G. T., Filho, G. P., Johnson, T.,
Carvalho, V., Pessin, G., ... & Ueyama, J. (2017). Assessing users’
emotion at interaction time: a multimodal approach with multiple
sensors. Soft Computing, 21, 5309-5323.
[16] D.-P. Sascha and M. Aus Berlin, “User Experience of Interaction with
Technical Systems.” Accessed: Mar. 21, 2024. [Online]. Available:
https://depositonce.tu-
berlin.de/bitstream/11303/2090/2/Dokument_1.pdf
[17] J.-C. Hong, M.-Y. Hwang, H.-F. Hsu, W.-T. Wong, and M.-Y. Chen,
“Applying the technology acceptance model in a study of the factors
affecting usage of the Taiwan digital archives system,” Computers &
Education, vol. 57, no. 3, pp. 2086–2094, Nov. 2011, doi:
10.1016/j.compedu.2011.04.011.
[18] M. A. Alqahtani, M. M. Alamri, A. M. Sayaf, and W. M. Al-Rahmi,
“Exploring student satisfaction and acceptance of e-learning
technologies in Saudi higher education,” Front. Psychol., vol. 13, p.
939336, Oct. 2022, doi: 10.3389/fpsyg.2022.939336.
8