The word “smart” has been used in various fields and is widely accepted to mean intelligence. Smart home service, one of the representative emerging technologies in the IoT era, has changed house equipment into being more intelligent, remote controllable, and interconnected. However, the intelligence and controllability of a smart home service are contradictory concepts, under certain aspects. In addition, the level of intelligence or controllability of a smart home service that users want may differ according to the user. As potential users of smart home services have diversified in recent years, providing the appropriate functions and features is critical to the diffusion of the service. Thus, this study examines the smart home service features that current users require and empirically evaluates the relationship between the critical factors and the adoption behavior with 216 samples from Korea. The moderating effect of personal characteristics on behavior is also tested. The results of the analysis provide various theoretical and practical implications.
We are experiencing a new era of Internet of Things (IoT), where many electronic devices surrounding us are interconnected by a network [
The global smart home market is expected to grow to USD 119.26 billion by 2022 [
Despite the long history and growing interest, smart home service has not been widely accepted. There are many reasons (e.g., high device prices, limited consumer demand, and long device replacement cycles) preventing smart home diffusion. The largest barrier is due to a lack of technology to establish the infrastructure of a smart home [
Thus, this research aims to find what users really want from smart home services and investigate which features affect a user’s intention to leverage the service. This study will define the user acceptance factors from a new perspective and present a theoretical model to verify precedent factors and outcomes. Through this empirical and behavioral analysis, it will determine the concept of “smart” as reflecting a user’s actual smart home needs and its major functional features. In order to confirm that the needs for smartization sought by people vary depending on their individual characteristics and environment, the research also studies how aspects including the type of housing, gender, age, and prior experience affect user intentions.
A smart home refers to a residence equipped with a communication network, high-tech household devices, appliances, and sensors that can be remotely accessed, monitored, and controlled and that provide services responding to the residents’ needs [
Evolution of smart home services.
Year | Phase | Technical background | Main function |
---|---|---|---|
1990s | Home automation | Broadband Internet | Household automation |
2000s | Home network | Smart phone and app | Remote monitoring & control |
2010s | Smart home | IoT & AI | Context awareness |
A smart home is an advanced form of traditional home automation. An early definition of a smart home, which was influenced by home automation, is using common communication devices to integrate with a variety of services at home, assuring economic, secure, and comfortable operation of the home [
Smart Home (
Recently, smart home services are evolving as they approach AI. The intelligent personal assistant “Alexa,” developed by Amazon Lab126, has been installed in a wide range of products. LG Electronics has adopted Alexa throughout its smart home product line. For example, if a user calls “Alexa” from a smart refrigerator, the user can access services such as searching news, online shopping, and checking schedules. In addition, China smart home manufacturer Xiaomi is planning to target the smart home market as part of its long-term vision. Xiaomi launched an air purifier that can be remotely controlled by a smart phone and developed a smart module that can be inserted into all appliances such as refrigerators, air conditioners, and washing machines. Apple is developing an AI speaker that supports “Apple HomeKit,” expected to provide voice support as a hub to control home kit products. Thus, smart home services are developing and proliferating by adopting IoT and AI.
Prior studies on smart homes are based on a technical or a partial approach. For instance, [
The word “smart” has been used in various fields such as smart phones, smart TVs, and smart learning, including smart homes. Although it has a slightly different meaning in each concept, it generally means “intelligent,” which can be interpreted as the concept of a weak level of artificial intelligence (AI). However, whether such a concept represents an intelligence service that can perfectly substitute the decision-making process of human beings requires debate. Humans in general will be reluctant to delegate all of the decision-making authority to machines considering their search for freedom, uncertainty, and distrust in technology. In addition, the level of smartness people demand would also vary depending on their individual characteristics and environment. Some people have vague fears about intelligent and smart things. For instance, when AlphaGo beat a human in the Go game, some people had negative perspectives on AI because a computer can control or be detrimental to people. Thus, the “smart” that people want may entail a limited scope of intelligence which is under the control of human beings, unlike the theoretical point of view.
Smart home service acceptance research has been active since the mid-2000s. Most studies have extended the technology acceptance model (TAM) or the unified theory of acceptance and use of technology (UTAUT) and have focused on specific groups such as the elderly, the disabled, and patients. Leeraphong et al. [
Recently, as smart services become more common, research on the acceptance of general users has become active. Bao et al. [
Some studies emphasize the importance of security and privacy in smart home acceptance. For instance, Tanwar et al. [
However, as the existing discussions are abstract, a fundamental understanding is necessary to characterize smart home services. In addition, past studies have conducted an empirical analysis on a specific group, but the popularization of smart home services now requires more general discussions for diverse user classes.
The initial smart home service was promoted through the automation of the domestic system, aiming for convenience, comfort, stability, amenity, health, reduction of household labor, and energy efficiency. Since then, developments of the wireless Internet and smart phones have extended the concept of a smart home to services that can be remotely controlled anytime and anywhere. In the IoT era, household electrical appliances and information and communication devices are interconnected, and the smart home is developing into a form of an artificial intelligence service that operates by self-understanding the behaviors of the residents. Therefore, the smart home in the IoT era is a concept that adds interconnectedness to the traditional characteristics of automation and remote controllability [
Automation is defined as the “execution by a machine agent (usually a computer) of a function that was previously carried out by human” [
The virtue of a smart home is that it can be controlled remotely by mobile devices. This is a core feature of a smart home system since users prefer to instantly control smart home services such as controlling lamps, curtains, and information appliances [
Interconnectedness is defined as the ability of devices, applications, and services to be connected with each other to work together [
Technical errors in integrated smart homes can be a concern to potential users. The reliability of smart home services depends not only on the fact that the technology will not malfunction but also on the fact that the technological components will function flawlessly while providing an accurate service [
The research model of this study was developed as shown in Figure
Research model.
Automation is a term referring to the automation of housework and household activities such as the control of lighting, heating systems, and ventilation. This kind of automation enables users to be comfortable, live conveniently, be secure, and be energy efficient. In addition, it monitors elderly and disabled people to ensure suitable care [
Controllability is the ability to do whatever a user needs with the given system that is under control [
Interconnectedness is defined as the ability to work together reliably owing to the fact that a discrete manufacturer exists [
Reliability between a manufacturer and a user is an important factor in user behavior. In the Maslow theory of human motivation safety, security and protection are the second needs to be satisfied after fulfilling basic physiological needs like food, water, and shelter [
Smart home services are no longer a specialized service for a specific group of people such as housewives, patients, or the elderly but are now developing into a more public service that the general public can use for a more convenient lifestyle. Such trends call for the need of a more detailed analysis of the actual motivations of diverse groups of users in order to generalize the research results and render them a reflection of the current times. As there could be a difference in the relationship between variables representing conventional users and new groups of users when considering their characteristics, the research investigated the differences between groups according to the type of housing, gender, age, and use experience, as moderator variables.
This study conducted an online survey in October 2015 to evaluate the research model. Data was collected by a professional survey company in Korea. The final 216 collected samples were used for analysis. The demographic distribution of the samples is as follows: 111 males (51.4%), 105 females (48.6%), 10s (28, 13%), 20s (37, 17.1%), 30s (45, 20.8%), 40s (46, 21.2%), 50s (38, 17.5%), and over 60 (22, 10.4%). The distribution of the samples was balanced for analysis (Table
Characteristics of the respondents.
Characteristics | Respondents ( | |
---|---|---|
Number | Percentage | |
Gender | ||
Male | 111 | 51.4 |
Female | 105 | 48.6 |
Age | ||
16–19 | 28 | 13.0 |
20–29 | 37 | 17.1 |
30–39 | 45 | 20.8 |
40–49 | 46 | 21.2 |
50–59 | 38 | 17.5 |
60+ | 22 | 10.4 |
Education | ||
Less than high school | 44 | 20.4 |
College or university | 150 | 69.4 |
Advanced degree | 22 | 10.2 |
Occupation | ||
Official worker | 60 | 27.8 |
Service worker | 8 | 3.7 |
Professional/researcher | 12 | 5.5 |
Self-employer | 14 | 6.5 |
Public service worker | 3 | 1.4 |
Student | 66 | 30.6 |
Housewife | 34 | 15.7 |
Other | 19 | 8.8 |
Residence type | ||
Apartment/multifamily house | 114 | 52.8 |
Single house | 102 | 47.2 |
Experience of smart home service | ||
Yes | 46 | 21.3 |
No | 170 | 78.7 |
All measurement items to measure latent constructs were developed based on previous studies. Responses were collected based on a 5-point Likert scale. The items of each construct are shown in Table
Survey items.
Construct | Item number | Measurement items | References |
---|---|---|---|
Perceived automation | PA1 | Smart home services help the residents proactively without human intervention. | [ |
PA2 | Smart home services provide autoadjusted control. | ||
Perceived controllability | PC1 | I can control every electrical device of smart home services through simple operation. | [ |
PC2 | It is convenient to control smart home services anywhere at any time. | ||
Perceived interconnectedness | PI1 | Smart home devices are interconnected with each other. | [ |
PI2 | Smart home services by integrating different device venders do not create problems. | ||
Perceived reliability | PR1 | I am not worried to use smart home services because other people or organizations may be able to access my account. | [ |
PR2 | There will not be much potential loss associated with disclosing personal information to the smart home service provider. | ||
PR3 | I think smart home service providers are reliable. | ||
Adoption intention | AI1 | Using smart home services is worthwhile. | [ |
AI2 | I intend to use smart home services in the future. | ||
AI3 | I predict I would use smart home services in the future. |
The reliability and validity of the constructs were checked. The reliability of latent variables can be confirmed by Cronbach’s
Validity of constructs.
Construct | Items | Factor loading | AVE (>0.5) | Composite reliability (>0.6) | Cronbach’s alpha (>0.7) | |
---|---|---|---|---|---|---|
Perceived automation | PA1 | 0.947 | 97.763 | 0.902 | 0.965 | 0.946 |
PA2 | 0.954 | 119.742 | ||||
Perceived controllability | PC1 | 0.898 | 61.407 | 0.903 | 0.949 | 0.893 |
PC2 | 0.831 | 23.021 | ||||
Perceived interconnectedness | PI1 | 0.935 | 94.090 | 0.748 | 0.855 | 0.667 |
PI2 | 0.902 | 38.725 | ||||
Perceived reliability | PR1 | 0.714 | 7.651 | 0.843 | 0.915 | 0.815 |
PR2 | 0.769 | 9.760 | ||||
PR3 | 0.769 | 12.624 | ||||
Adoption intention | CONT1 | 0.947 | 114.355 | 0.564 | 0.795 | 0.647 |
CONT2 | 0.964 | 163.208 | ||||
CONT3 | 0.938 | 101.322 |
Correlations of the constructs and square root of AVE.
PA | PC | PI | PR | AI | |
---|---|---|---|---|---|
PA | |||||
PC | 0.68 | ||||
PI | 0.54 | 0.78 | |||
PR | 0.36 | 0.56 | 0.51 | ||
AI | 0.50 | 0.66 | 0.62 | 0.49 |
Construct cross-loadings.
PA | PC | PI | PR | AI | |
---|---|---|---|---|---|
PA1 | 0.637 | 0.492 | 0.355 | 0.457 | |
PA2 | 0.650 | 0.540 | 0.334 | 0.486 | |
PC1 | 0.716 | 0.615 | 0.513 | 0.629 | |
PC2 | 0.426 | 0.757 | 0.448 | 0.497 | |
PI1 | 0.543 | 0.735 | 0.480 | 0.616 | |
PI2 | 0.448 | 0.696 | 0.453 | 0.507 | |
PR1 | 0.162 | 0.321 | 0.303 | 0.249 | |
PR2 | 0.188 | 0.285 | 0.301 | 0.308 | |
PR3 | 0.389 | 0.564 | 0.482 | 0.472 | |
AI1 | 0.446 | 0.581 | 0.553 | 0.437 | |
AI2 | 0.482 | 0.662 | 0.598 | 0.463 | |
AI3 | 0.484 | 0.625 | 0.602 | 0.490 |
Structural equation modelling results are presented in Figure
Results of the structural model:
However, the results were very diverse when analyzed by grouping according to the residence type, gender, age, and experience (Table
Result of path coefficients by grouping.
Residence type ( |
Gender | Age | Experience | |||||
---|---|---|---|---|---|---|---|---|
APT (114) | House (102) | M (111) | F (105) | <39 (110) | 40+ (106) | Yes (46) | No (170) | |
H1 | −0.053 | 0.271 |
0.153 | 0.004 | 0.015 | 0.159 | 0.332 |
0.047 |
H2 | 0.367 |
0.317 |
0.291 |
0.378 |
0.256 |
0.378 |
0.264 | 0.310 |
H3 | 0.301 |
0.129 | 0.251 |
0.217 | 0.337 |
0.134 | 0.062 | 0.272 |
H4 | 0.148 | 0.165 |
0.142 | 0.184 |
0.157 | 0.177 |
0.342 |
0.140 |
The objective of this study was to understand and explain customers’ behavioral intentions to adopt smart home services. Unlike the previous research that examined the user behavior associated with smart home service adoption based on acceptance theories, this study captures the characteristics of smart home services and presents a new theory and model. The empirical analysis of the proposed research model demonstrated various implications.
In all samples, three factors: controllability, interconnectedness, and reliability had a significant impact on the acceptance behavior of a smart home service. It is very interesting that automation does not have a significant effect. This can be interpreted as follows: people generally seek relatively safer and more effective remote management features rather than highly advanced automated services. People may want the devices of a smart home to be under their control rather than being fully automated because a home is safe and represents their personal space where they can rest. Considering that controllability is the most important antecedent for adoption, it becomes apparent that the automation that people want is eventually intelligent and represents an optimal controllability that is close to a limited form of automation.
There are also various findings in group comparison analysis. Controllability and interconnectedness significantly affect adoption for people living in apartments while automation and reliability are considered to be important for general home residents. The difference in the level of infrastructure between apartments and general homes may be an influence. Recently, new apartments in Korea have been provided with smart home services (remote heating management, gas shutdown, etc.). As a result, apartment residents seem to want more extensive and precise control. As general homes do not provide any networked and automated functions to control households, general home residents may want the automation and reliability of a smart home service.
When comparing gender, men emphasize interconnectivity while women emphasize reliability. Similar results were observed in the age-related comparisons. This can be attributed to the risk avoiding tendency of women and older people. It is understandable that they prefer these factors because they seek stability compared to men or young people. Men and young people tend to prefer interesting and innovative services, and the interconnectedness of the smart home can serve as a factor in meeting these needs.
Those who have experienced similar services emphasize automation and reliability. In fact, it is likely that those who do not place much significance on control and interconnection have experienced that the control and connectivity of past services did not guarantee usability.
This study makes several contributions to theory. First, it presents the specific success factors of smart home service adoption and empirically analyzes the relationship with the acceptance behavior. Most studies have presented abstract success factors (e.g., usefulness and enjoyment) based on technology acceptance theory, but this study derived the detailed critical factors through a literature review. Second, this study captures the concept of the word “smart” in the smart home service. According to the result, the smart that people desire is close to intelligent control but not fully automated. This explains why past services (e.g., home automation and the networked home) that are similar to a smart home have not spread. The past services have not been able to meet the desired level of automation for users. Third, the results of the comparison between the groups show that the factors that influence smart home adoption can be different according to the characteristics of users in the case of a smart home service. In the case of a smart home service, the main users are often the elderly, patients, and women, and the results clearly show the effect of user characteristics on acceptance behavior. Therefore, acceptance studies that may be different according to user characteristics demonstrate that a research design needs to consider user characteristics.
This study also provides several useful insights for practitioners who manage the development or marketing of a smart home service. First, functional diversity should be assured to consumers. Due to the different smart home service requirements determined by the customer group, it is necessary to systematically support each consumer group in selecting the desired functions. For example, it is possible to provide a smart home service with high interconnectivity with other devices for apartment residents, while focusing on delivering an automation function for general home residents. To this end, smart home service companies should consider how to configure smart home services, including cooperating with third party device manufacturers for each type of customer, and prepare various plans according to the detailed function configuration. Second, continuous R&D on AI-based automation is required. It was revealed that automation has a positive impact on the intention for continuous use of current smart home users. Therefore, it is reasonable to infer that as the basic controllability-based smart home service spreads, consumers’ need for automation will increase. A large-scale investment on basic infrastructure such as data centers, cloud, or big data systems by smart home service companies will be required in order to analyze customers’ lifestyles and interact with their movements of emotional change. Third, to increase customer reliability, smart home service providers should adopt high-level security technologies and set up internal policies to prevent information leakage. Trust in smart home service providers has become a significant issue as data-based smart home companies are rapidly expanding, such as Google.
This study empirically examined important factors for the adoption and spread of smart home services. Research results show that interconnectivity and reliability are required along with the right level of automation. Furthermore, because there are differences in preference factors according to the characteristics of users, it has been confirmed that the service design that considers these characteristics is necessary. If these factors are taken into consideration, smart home services that have not been activated in the past will spread and the market will grow.
Although the findings of this study provide meaningful insights into the adoption of smart home services, this study has limitations that future research should address. First, key findings of this research are based on the data only from South Korea. A future study should attempt to gather ethnically and geographically diverse data to ensure the generalizability of the results. Second, hedonic-related variables may be considered as influence factors of adoption intention such as perceived design (e.g., visual attractiveness of control hub hardware and software user interface). Lastly, in future studies, a significant difference in the antecedents’ influence on behavioral intention between current and potential users may be found. Despite the limitations, this study contributes to a more systematic understanding of smart home service adoption. In this regard, we hope that this study helps to create a foundation for related future research.
Authors declare that there is no conflict of interest regarding the publication of this paper.
This study was supported by the BK21+Intellectual Property·Information Protection Law Expert Program.