Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

Miaojia Lu; Chengyuan Huang; Ran Wang; Hao Li

doi:10.1155/2023/3440691

Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

Lu, Miaojia;Huang, Chengyuan;Wang, Ran;Li, Hao 2023-03-08 00:00:00 Hindawi Journal of Advanced Transportation Volume 2023, Article ID 3440691, 17 pages https://doi.org/10.1155/2023/3440691 Research Article Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience 1 1 2 1 Miaojia Lu , Chengyuan Huang , Ran Wang , and Hao Li College of Transportation Engineering, Tongji University, Shanghai 201804, China College of Civil Engineering, Hunan University, Changsha 410082, China Correspondence should be addressed to Chengyuan Huang; hcyuann@163.com Received 16 September 2022; Revised 5 February 2023; Accepted 17 February 2023; Published 8 March 2023 Academic Editor: Michela Le Pira Copyright © 2023 Miaojia Lu et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Te high demand and acute timeliness that characterizes instant delivery entail the challenges of high labor costs and an increase in courier trafc accidents. Autonomous delivery vehicles (ADVs) may serve as a key solution, with their attendant reduced labor input and higher efciency. Customers play a key role in the successful implementation of ADVs on a large scale. However, understanding the factors that afect customers’ intentions to use ADVs is still limited. Compared to autonomous driving, ADV customers are ultimately not the real users, who only are served by ADVs during the last leg of a trip. On account of this, the Technology Acceptance Model (TAM) may not be well-ftted for explaining the dynamics involved in ADV adoption. Within the context of ADVs, our study identifed infuencing factors that have not been captured by prior studies. Tis study incorporates infection risk, use experience, and social awkwardness into the Difusion of Innovation (DOI) theory to explore customers’ intentions to use ADVs. Data from 691 survey respondents were collected to validate the research design. Te results demonstrate that compatibility, social infuence, infection risk, green image, social awkwardness, and use experience all have a signifcantly positive impact on customers’ intentions to adopt ADV services, while complexity and perceived risk both exhibited a negative impact. But no efect could be found for relative advantage, which may be because of the fact that customers only need ADVs to meet their delivery demand. Tis study contributes to understanding customers’ adoption intentions toward ADVs, informing policymakers in formulating ADV regulations and standards, and promoting the large-scale application of ADVs in instant delivery services. accidents related to instant delivery. In August 2020, ap- 1. Introduction proximately 12 thousand trafc violations related to instant With the acceleration of the pace of life, combined with the delivery occurred in Shenzhen, China [4]. In addition, the ease and convenience of online shopping and increases in working-age population in China has been declining since income, more and more people have chosen to use instant 2013 and nearly 25% of couriers work more than 12 hours delivery services [1]. Instant delivery orders include not only a day [5]. Te concomitant rising human labor costs account meals but also fresh food, fowers, and other products such for 50 to 60% of the total cost of last-mile delivery [6]. as medicine and important documents. Te scale of instant Terefore, many startup companies and Internet companies delivery reached 24.37 billion orders in China in the year such as Amazon, Nuro, Jingdong, and Meituan [7] have 2020, a ten-fold increase since 2015 [2]. In particular, the begun to explore innovative delivery technologies such as frequency of customers shopping online for fresh food grew Autonomous Delivery Vehicles (ADVs), which are self- 71.2% during the COVID-19 pandemic [3]. Te increasing driving vehicles that travel on roads (as distinct from order volume, the importance of freshness, and the demand other autonomous robots that travel on sidewalks) to deliver for timeliness have consequently led to more courier trafc goods. ADVs can reduce labor costs and accident rates, as 2 Journal of Advanced Transportation intentions can aid in advancing large-scale application of well as increase delivery efciency, as one ADV can replace two couriers [8]. In the United States, Europe, and China, ADV technology. In the meantime, legislators are tackling how to properly regulate the use of ADVs. Te fndings of many regulations have been enacted to support the public road operation of ADVs, with corresponding operation this study as they relate to customer behaviors can also assist requirements [7, 9, 10]. Due to the often intensely chal- in the formulation of sound legislation. lenging conditions that have arisen from the COVID-19 Te remainder of the study is organized as follows: pandemic, customers, businesses, and governments have Section 2 reviews previous studies about the acceptance of switched from being cautious beta testers to becoming eager autonomous vehicles (AVs), ADVs, unmanned aerial ve- early adopters [11]. Te novelty, convenience, efciency, and hicles (UAVs), and autonomous robots, as well as studies using DOI theory. Section 3 details the constructs and contactless delivery associated with ADVs have led cus- tomers to develop high expectations for this technology [12]. hypotheses in the proposed theoretical model. Tis section also describes the questionnaire designed to investigate However, many customers are still skeptical and cau- tious about the viability of ADVs. Accidents involving au- potential customers’ intentions and the methods used to analyze the collected data. Section 4 presents the descriptive tonomous vehicles are worrisome to the public. In October 2021, a trafc accident involving an ADV and a private car statistics and discusses the results of the model assessments. took place in Beijing, with the ADV found to be primarily Section 5 recaps some key contributions of this study, in- responsible for the accident [13]. On August 14, 2021, an dicates limitations of the study, and suggests directions for entrepreneur died while driving NIO’s ES8 SUV when the future research. automatic cruise system was turned on. In addition, ADVs may break down on the way or lose customers’ goods [1, 14]. 2. Literature Review Moreover, the massive amount of data required for oper- ating autonomous vehicles has raised concerns about pri- 2.1. Factors and Teories Related to Autonomous Vehicles vacy. Some have expressed alarm about the risk of hackers Acceptance and Intentions. Here, we review previous studies infltrating ADV systems and controlling ADVs to attack related to acceptance and intentions toward autonomous pedestrians [15]. It has been argued that the biggest barrier vehicles, with particular attention to autonomous driving to widespread adoption of ADVs is not technological ca- and autonomous delivery. Teoretical models used in pre- pability but social acceptance. Customers play a key role in vious research are shown in Table 1. the successful implementation of ADVs on a large scale [16]. Terefore, factors that may have signifcant infuences on customers’ adoption intentions need to be explored to ex- 2.1.1. Autonomous Driving. Autonomous driving refers to pand ADV acceptance as well as to integrate their consid- cars capable of sensing its environment and operating eration in the development of company logistics and without human involvement. Many researchers have in- government regulations. Te aim of this study is to in- vestigated the acceptance of autonomous driving. Some vestigate factors that signifcantly infuence customers’ in- studies have been conducted based on behavioral theories tentions to use ADV, drawing upon behavior theories. [18, 21, 28, 34], while other studies have investigated par- Difusion of Innovation (DOI) theory is applied in this ticular factors to explain user acceptance [36–41]. study because of the following reasons: (1) it can be used to Perceived usefulness and perceived ease of use are both explore the characteristics of personal innovativeness and important factors proven to afect user acceptance and in- (2) this theory is suitable for examining the customers, who tention [18, 19, 42]. Many researchers add other factors such are not the real users of ADVs [17] but are served by ADVs as green image [18] and perceived trust [19, 22] into the during the last leg of a trip. Our study also identifes factors Technology Acceptance Model (TAM) for further analytical afecting adoption that have not been captured in previous refnement. Using the Unifed Teory of Acceptance and Use studies. ADVs have played an important role in the delivery of Technology (UTAUT), Adnan et al. [27], Hewitt et al. of daily necessities and medicine during the COVID-19 [28], and Meyer_Waaeden and Cloarec [29] all investigated pandemic because of their ability to facilitate contactless the importance of performance expectancy, efort expec- delivery. Given this key characteristic of ADVs and the tancy, trust, and social infuence in AV adoption. Liu et al. ongoing public health context in which ADVs operate, [36] studied customers’ willingness to pay for self-driving infection risk is considered in this study. Because ADVs can vehicles and they found that more than 50% of respondents bypass person-to-person contact, another factor, i.e., social were willing to pay extra, although the amounts are diferent. awkwardness, is also considered. Use experience of ADVs is In addition, Berrada et al. [43] divided customers into fve also incorporated into the research design. Tis study is the clusters including conservatives, skeptics, late adopters, early frst to analyze the adoption intention of customers with adopters, and explorers and found more than 60% of re- regard to ADVs based on DOI theory. Tis study also spondents were willing to use autonomous vehicles. Similar contributes to existing research by considering infection to this study, Bernhard et al. [41] conducted two subsurveys risk, use experience, and social awkwardness in relation to to investigate the efects of autonomous minibus using ADV adoption intentions. Gaining a more comprehensive experience and found that respondents who already had understanding of the elements that infuence adoption experience using autonomous minibuses were more willing Journal of Advanced Transportation 3 Table 1: Teory models applied in previous studies of autonomous driving and autonomous delivery. Teory Teory description References AVs: Wu et al. [18]; Liu et al. [19]; Dirsehan and Can [20]; Chan TAM is used to explain how perceived usefulness and perceived and Lee [21]; Li et al. [22] Technology acceptance model (TAM) ease of use afect personal behaviors [18] UAVs: Leon et al. [23]; Michels et al. [24]; Choe et al. [25] Autonomous robots: Abrams et al. [26] UTAUT was developed from TAM, TPB, and other models to AVs: Chan and Lee [21]; Adnan et al. [27]; Hewitt et al. [28]; Unifed theory of acceptance and use of technology explore user acceptance and includes four constructs (i.e., Meyer-Waaeden and Cloarec [29]; Zhang et al. [30] model (UTAUT) performance expectancy, efort expectancy, social infuence, UAVs: Kim and Chung [31]; Ganjipour and Edrisi [32] and facilitating conditions) [21] ADVs: Kapser and Abdelrahman [16]; Kapser et al. [33] CAT holds that emotions are aroused due to the individual’s Cognitive appraisal theory (CAT) evaluation of the stimulus, which further determines the AVs: Ribeiro et al. [34] behavioral response [34] MCI emphasizes the efect of motivation on customers’ AVs: Hewitt et al. [28] behaviors. Motivation is classifed into four main types, namely, Motivated consumer innovativeness (MCI) functional motivation, hedonical motivation, cognitive UAVs: Mathew et al. [1] motivation, and social motivation [1] TPB is designed to predict customers’ behavioral intentions Teory of planned behavior (TPB) according to three main components, namely, attitude, UAVs: Ramadan et al. [35] subjective norms, and perceived behavioral control [35] DOI is used to explain why customers adopt new technologies Difusion of innovation theory (DOI) and includes fve constructs (i.e., relative advantage, UAVs: Yoo et al. [14] complexity, compatibility, observability, and trialability) [14] 4 Journal of Advanced Transportation respondents showed a willingness to pay more, with one to accept the technology because respondent confdence levels often increase with such direct experience. group willing to pay up to $2.92 (average) or between 0-$6 for the service. Abrams et al. [26] proposed a new concept, “existence acceptance” to extend TAM. Existence accep- 2.1.2. Autonomous Delivery. Autonomous delivery includes tance refers to pedestrians’ or other trafc respondents’ not only ADVs but also UAVs and autonomous robots. acceptance of sharing the same roads with autonomous UAVs, also commonly referred to as drones, are fying robots. robots that can be remotely controlled or autonomously fown to deliver more light-weight, compact goods. Au- tonomous robots refer to robots that also operate and travel 2.1.3. Teoretical Background. In previous studies of au- upon sidewalks, usually complemented by a mothership van tonomous driving and autonomous delivery, TAM and that can transport the robots close to the delivery zone [7]. UTAUT are the most widely used behavioral theories for exploring new technology adoption (see Table 1). However, (1) UAVs. Combining TAM and DOI theory, Yoo et al. [14] TAM and UTAUT are not suitable for explaining customer adoption intention toward ADV services because customers proposed a model to investigate the factors that afect customer perceptions of UAVs and analyze the efect of are not the real users of ADVs. Te logistics practitioners are delivery area. Te researchers found that two major per- the direct users of ADV technology, not the customers, who ceived advantages of UAVs are their speed and environ- do not need to be familiar with the technology [17]. Te mental friendliness. In their study, perceived risk was also perceived usefulness and perceived ease of use are not the incorporated and defned to include performance risk, de- main concerns of customers when deciding whether to livery risk, and privacy risk. Based on customer motivated adopt ADVs in instant delivery. Tus, TAM and UTAUT innovativeness, Hwang et al. [44] applied consumer in- may be not well suited for examining and characterizing novativeness theory to study customer attitudes and ADV adoption intention. Another approach that could be considered to investigate technology adoption is Cognitive adoption intentions toward UAV food delivery. Te results showed that attitude played an important role in intention to Appraisal Teory (CAT). Tis theory holds that emotion is a kind of individual mental state that is generated when use and intention to pay more. Mathew et al. [1] explored customer intention to adopt UAVs for food delivery in evaluating relevant information and that infuences indi- India. Perceived risk and green image were found to be vidual behavioral motivation. One advantage of CAT is that signifcant predictors of customer attitude and intention. it can analyze customer behavior through a wide range of Aydin [45] investigated public knowledge and acceptance of emotional experiences. But at the same time, too many UAVs in specifc application areas, for example, commercial cognitive evaluation dimensions make it difcult for re- and public safety applications. However, the results showed searchers to capture quick and unconscious evaluation re- sults. Tis theory also does not answer the question that that UAVs were not generally accepted except for in the arenas of public safety and scientifc research. Need for whether any kind of emotion can be acquired without an evaluation system [48]. Motivated consumer innovativeness human interaction was added into the UTAUT2 model in Ganjipour and Edrisi’s study and was found to have (MCI) theory models motivation according to four main dimensions, namely, (1) functional, (2) hedonical, (3) cog- a negative efect on adoption intentions toward delivery drones [32]. Other studies used the Knowledge, Attitude, nitive, and (4) social [1]. In MCI, social motivation refers to and Practice Model or other methods to explore customer customers wanting to improve their image by purchasing acceptance [45–47]. innovative products/services (rather than referring to cus- tomers imitating the purchasing decisions of those in their (2) ADVs. Kapser and Abdelrahman [16] studied customer social networks). Teory of Planned Behavior (TPB), which intention toward ADVs and analyzed various key behavioral assumes that individuals consciously control their actions, has been used to study a wide range of health-related be- components (e.g., performance expectancy, efort expec- tancy, perceived risk, and price sensitivity) based on UTAUT haviors and learning behaviors. It includes three main variables, i.e., attitude, subjective norms, and perceived theory. Te results showed that respondents were neutral on the use of ADVs as a delivery alternative while price sen- behavior control. Although this theory has been applied by sitivity exerted a great infuence on customer intention. many researchers, there is a lack of agreement about its main Building on this research, Kapser et al. [33] integrated trust variables, raising doubts about the accuracy of results from and innovativeness into their acceptance model while factors research based on this theory. In addition, there are many shown to be insignifcant (i.e., efort expectancy) were re- behaviors that may not be explained by the main variables moved. In addition, a moderating factor, i.e., gender, was profered for this theory [49]. As an alternative, DOI theory explains how innovative technologies and ideas spread applied and they found that social infuence, perceived risk, and hedonic motivation had particularly signifcant efects through society over time and is also applied in commu- nication and information systems research to help explain on the use intentions of women customers. why customers adopt innovative technologies [14, 50]. Given its particular focus on innovative technology adoption, DOI (3) Autonomous robots. Pani et al. [11] analyzed customers’ willingness to pay more for autonomous robots by dividing theory is selected as the behavior theory of this study customer groups based on latent class analysis. 61.28% of of ADVs. Journal of Advanced Transportation 5 Table 2: Constructs used in DOI. Construct Main fndings References Duan et al. [62]; Ooi et al. [55]; Wang et al. [56]; Chung [57]; Kang et al. [58]; Relative advantage Relative advantage can promote adoption intention Hamad et al. [63]; Yoo et al. [14]; Min et al. [60] Complexity Complexity is a barrier to adoption intention Duan et al. [62]; Wang et al. [56]; Chung [57]; Yoo et al. [14]; Min et al. [60] Duan et al. [62]; Wang et al. [56]; Chung [57]; Kang et al. [58]; Hamad et al. [63] Compatibility Compatibility can promote adoption intention Min et al. [60]; Yoo et al. [14] Observability Observability can promote adoption intention Duan et al. [62]; Chung [57] Trialability Trialability can promote adoption intention Duan et al. [62]; Chung [57]; Hamad et al. [63] Social infuence Social infuence can promote adoption intention Min et al. [60] Green image/environmental Green image/environmental concern can promote adoption Yoo et al. [14] concern intention Perceived risk Perceived risk is a barrier to adoption intention Yoo et al. [14] 6 Journal of Advanced Transportation adoption intentions toward ADVs in China by adding the 2.2. DOI Teory Models of User Perception toward a New Technology. DOI theory as presented by Rogers [51] em- infuences of use experience, infection risk, and social awkwardness to DOI theory. ploys more specifc innovation characteristics, which are not typically found in behavioral theories in order to explain the adoption of new technologies or products. According to 3. Methodology Rogers, “the innovation is an idea, practice, or object that is In this section the constructs that are used in this study’s perceived as new by an individual or other unit of adoption.” customer adoption intention model are identifed and the Rogers also contends that innovations are difused through model structure is described. Customer adoption in- various channels and within a particular social system. tention here refers to the degree of willingness to use Rogers [52] presented fve innovation characteristics, ADVs (i.e., intention to use) and the willingness to pay namely, (1) relative advantage, (2) complexity, (3) com- more for the use of ADVs (i.e., intention to pay more). patibility, (4) observability, and (5) trialability. Relative First, the conceptual research model, which includes DOI advantage is defned as “the degree to which an innovation is theory supplemented with three new constructs, is pre- perceived as better than the idea it supersedes.” Complexity sented. Te proposed “Customer Adoption Intention is defned as “the degree to which an innovation is perceived toward Autonomous Delivery Vehicle (CAI-ADV)” as difcult to understand and use.” Compatibility involves model is depicted in Figure 1. Next, the survey design is “the degree to which a service is perceived as consistent with delineated, including the time, place, and scale of the users’ existing values, beliefs, habits, and present and pre- questionnaire, as well as the questionnaire’s content. vious experiences.” Observability is defned as “the degree to Finally, the data analysis process is presented. In addition which the innovation can be visible to others.” Finally, to the constructs, demographic factors as control vari- trialability is defned as “the extent to which the innovation ables are considered in our analyses. can be trialed, modifed, and experienced before the adoption” [53]. However, previous studies found that only three characteristics—relative advantage, complexity, and 3.1. Constructs in the Conceptual Research Model and Pro- compatibility—impact innovation adoption [14, 50, 54–62]. posed Hypotheses. Te constructs are divided into two parts, DOI theory has proven to be an efective theory for exploring namely, the original constructs from DOI and constructs customers’ acceptance of a new technology. Constructs used that we added. Te former includes relative advantage, in the original formulation of DOI theory and added in other complexity, and compatibility. Te latter encompass social subsequent studies are reviewed in Table 2. As indicated in infuence, perceived risk, green image, infection risk, social Table 2, relative advantage, complexity and compatibility are awkwardness, and use experience. Defnitions, related re- the most frequently applied constructs of DOI. Social in- search, and our hypotheses for each construct are introduced fuence has moreover been found to promote the adoption of as follows: new technologies [60]. Green image and perceived risk have Relative advantage (RA) is defned as “the degree to also exhibited positive efects on adoption intention which an innovation is perceived as better than the idea [1, 14, 18]. it supersedes” [53]. Previous research has found RA to be a positive predictor of behavioral/use intention [14, 55–58, 60, 62, 63]. In this study, relative advantage is 2.3. Research Gap. A review of the literature in the auton- defned as a customer’s subjective evaluation of the omous driving and autonomous delivery felds revealed only benefts aforded by ADVs against those by courier two studies [16, 33] that appear to explicitly concentrate on delivery. Autonomous delivery typically enables cus- the acceptance of ADVs. In contrast with these two studies, tomers to schedule delivery of their orders at any a time which were based on UTAUT theory, the DOI theory of their choice [64]. Customers can place orders at employed in this study focuses more on how innovative midnight and be served by ADVs without worrying technologies spread in society rather than on the use of those about order rejection. Time convenience is an im- technologies. We also incorporate some new constructs that portant characteristic of ADVs, making it a particularly may afect the ADV adoption intention of end customers, attractive delivery option for customers. Moreover, such as use experience and infection risk. As ADVs have without the personal preferences of couriers hampering become regularly available as a delivery option in Anting, options, orders from remote areas would not be Shanghai, some respondents to our survey have already used rejected and wait times may be greatly reduced during ADVs, with their use experiences increasing their confdence inclement weather. Time convenience [58] and zero in ADV services. Te risk of COVID-19 infection caused by rejection rates in harsh conditions may result in more person-to-person contact is likely to encourage customers to adoption. In addition, there are reports of couriers more readily adopt ADV supported services due to the contaminating customer food orders [65]. In contrast contactless delivery that ADVs enables. Tis study also to courier delivery, delivery by ADVs could better considers the impacts of social awkwardness on ADV protect the cleanliness and integrity of goods. Tere- adoption intention as ADVs can bypass direct communi- fore, the following hypothesis is proposed: cations with couriers and may become more popular among customers who are not so adept at socializing. To the best of H1: Relative advantage is positively related to customer our knowledge, this study is the frst to investigate customer intentions toward using ADV services. Journal of Advanced Transportation 7 Compatibility Relative advantage Complexity H2 H3 H1 Social influence Infection risk H7 H4 Customer H5 H8 adoption Perceived risk Social awkwardness intention H6 H9 Use experience Green image Income Education Age Gender Constructs from DOI Added Constructs Control variables Figure 1: Customer adoption intention toward autonomous delivery vehicle model (CAI-ADV). Complexity is defned as “the degree to which an in- home. Such an alteration engendered by the new de- livery mode demonstrates a gap in compatibility as novation is perceived as difcult to understand and use” [53]. Duan et al. [62], Wang et al. [56], Chung [57], compared to the customer’s usual delivery demand. As Yoo et al. [14], and Min et al. [60] found that com- a result, the customer’s adoption intention is lower. On plexity had a negative infuence on adoption of a new the other hand, if ADV service is not much diferent technology. In our study, complexity refers to the from courier delivery experiences, such as the ADVs problems that may arise in relation to a customer delivery price being the same as the previous delivery getting a notifcation when an order has arrived and/or price, customers’ ADV adoption intention would be difculties possibly experienced by a customer in re- higher. Terefore, the following hypothesis is proposed: trieving the food from the vehicle. For example, the H3: Compatibility is positively related to customer operation interface is challenging for some elderly intentions toward using ADV services. people to use, especially for picking up goods through Social infuence can be described as the phenomenon in a verifcation code. Without person-to-person contact which “people adopt when enough other people in the with a courier, it can be more difcult to return and group have adopted, that is, innovations spread by exchange goods. Terefore, the following hypothesis is a conformity motive” [66]. In other words, people will proposed: consciously imitate behaviors as exhibited in their H2: Complexity is negatively related to customer in- social network. Consistent with Higgins et al. [67], Min tention toward using ADV services. et al. [60] also found that social infuence had a positive efect on the acceptance of a new technology. Social Compatibility is defned as “the degree to which infuence is also considered in this study in order to a service is perceived as consistent with users” existing explore the positive feedback and recommendations values, beliefs, habits, and present and previous ex- communicated between members of close social groups periences” [53]. Studies from Duan et al. [62], Wang et as they relate to intentions toward ADV adoption. al. [56], Chung [57], Kang et al. [58], Hamad et al. [63], Terefore, the following hypothesis is proposed: and Min et al. [60] show that compatibility can posi- H4: Social infuence is positively related to customer tively afect customers’ intentions. Whether ADV intentions toward using ADV services. service is compatible with customers’ usual delivery demands will afect their intentions toward using Perceived risk has been generally found to be a useful ADVs. In this study, compatibility refers to the con- addition to DOI in studying customer intentions to- sistency between the new ADV service and some ward a new technology. Yoo et al. [14] and Mathew et previous experience with or more habitual use of in- al. [1] both used three dimensions to characterize perceived risk, performance risk, delivery risk, and stant delivery by customers. For example, ADVs cannot provide home delivery services in some residential privacy risk. ADV technology has the potential to al- leviate the problem of courier accidents. Whether areas or ofce buildings. A customer that used to enjoy door-to-door delivery service may then need to pick up ADVs can complete the delivery process and avoid goods delivered by ADVs in a certain area such as trafc accidents (i.e., delivery risk) is a thought-pro- a parcel locker that is three hundred meters away from voking problem. ADVs have not been implemented 8 Journal of Advanced Transportation the less intention there is to try drone delivery [32]. extensively enough to accurately evaluate their ability to reduce accidents. Due to the need to draw on a large Drawing upon this fnding, customers who are socially awkward, meaning that they are less adept at or in- amount of sensitive data online such as names, ad- dresses, and phone numbers, protecting privacy is clined to communicate with other people, may fnd another important concern that may afect customer ADVs to be an attractive option. Terefore, the fol- intentions. Furthermore, network instability can also lowing hypothesis is proposed: afect the operation of ADVs. An additional technical H8: Social awkwardness is positively related to cus- risk is hacking (e.g., breach of frewalls). Considering tomer intentions toward using ADV services. these risks, the more they are reduced, the higher the Studies from Bernhard et al. [41] show that use ex- adoption intentions of customers. Previous studies perience can afect customers’ intentions because found that perceived risk had a negative efect on ac- customers respond on the basis of real experience, not ceptance of a new technology [1, 14]. Terefore, the imagination. However, few studies have taken this perceived risk discussed in this study refers to the construct into consideration. Customers who have used delivery risk, privacy risk, and technical risk. In light of ADVs are better able to understand ADV delivery these considerations, the following hypothesis is service, including the process, its advantages, and proposed: disadvantages. Customer confdence in the service H5: Perceived risk is negatively related to customer could then increase as a result of direct experience. intentions toward using ADV services. Based on previous experiences, customers may be more likely to choose ADVs as a delivery option. Terefore, Because of global warming and increasing air pollution, the following hypothesis is proposed: the environmental impacts of any given transportation system has become a factor worth considering [68]. Te H9: Use experience is positively related to customer emergence of autonomous vehicles powered by elec- intentions toward using ADV services tricity enables cleaner energy consumption [64]. Te implementation of ADVs is expected to be a valuable step towards zero-emission logistics in the United 3.2. Survey Design. Based on the research model proposed in States [69]. Many studies show that customers’ envi- Section 3.1, we designed and conducted an online and ofine ronmental concerns could afect their adoption in- survey to test the proposed hypotheses. Our online survey tentions toward green products such as Avs and UAVs was conducted through the Sojump website, and the ofine [1, 14, 18]. In this study, green image is defned as survey was conducted in Anting, Shanghai. In Anting, the “customers” perceptions of ADVs having a salutary Yonghui supermarket began implementing ADV service in environmental performance.” Terefore, the following 2020, covering a delivery area of 5 square kilometers with hypothesis is proposed: more than six thousand households [71]. Te size of a sample from a given population necessary to H6: Green image is positively related to customer in- achieve a good or very good degree of adequate represen- tentions toward using ADV services. tativeness has been suggested to be 300 or 500, respectively Most countries around the world have been greatly [72] or alternatively achieved through a minimum 20 :1 ratio afected by COVID-19. Te disease spreads primarily between the sample size and the number of model pa- through contact with an infected person (even through rameters to be statistically estimated [73]. Tus, the 691 valid a cough or sneeze) or touching a surface that has the samples collected for our survey data analysis satisfy both of virus on it. Due to the fear of infection, contactless these recommendations. Te data collection took place in delivery (as enabled through ADVs and UAVs) has October and November in 2021. become increasingly popular, especially in areas ex- An introduction to ADVs is presented at the beginning periencing high infection risk. Te chance of human- of the survey, including history, delivery process, equipment to-human transmission during the COVID-19 pan- parameters, advantages, and disadvantages, in order to demic is reduced through the implementation of provide a clearer understanding of ADVs for respondents. ADVs. Terefore, the following hypothesis is proposed: Te second part of the survey includes questions about H7: Infection risk is positively related to customer respondents’ intention to adopt ADV services. Tese intentions toward using ADV services. questions are developed with reference to previous relevant Te results of Xi et al.’s study [70] show that same-day- literature. Consistent with previous surveys [1, 16, 26, 44], delivery has a negative efect on greeting/chatting with constructs except for use experience are measured based on neighbors but has a positive association with social a seven-point Likert-type scale, ranging from strongly dis- gatherings that often occur between acquaintances. agree (1) to strongly agree (7). Based on the hypotheses ADV technology can make our lives more convenient proposed in 3.1, the related observational variables and but can also further alienate us from our neighbors or corresponding references are presented in Table 3. Te third other casual acquaintances. Similar to autonomous part of the survey includes basic individual information (e.g., taxis [22], ADVs circumvent the person-to-person age, gender, monthly household disposable income, edu- contact that comes with using couriers. One study has cation, job, and instant delivery experience). To improve the found that the greater the need for human interaction, validity of the questionnaire, reverse worded questions are Journal of Advanced Transportation 9 Table 3: Constructs, observational variables, and supporting references. Construct Item Source RA1: Using autonomous delivery vehicles would help me receive orders more quickly Relative advantage (RA) RA2: Autonomous delivery vehicles can operate in inclement weather Mathew et al. [1] RA3: Autonomous delivery vehicles can deliver at any time I want RA4: Using autonomous delivery vehicles would be cool COL1: Interacting with autonomous delivery vehicles is clear and understandable COL2: I believe that it is easy to make autonomous delivery vehicles to do what I Complexity (COL) Yoo et al. [14]; Wu et al. [18] want COL3: I believe that retrieving goods from autonomous delivery vehicles is easy COA1: Using autonomous delivery vehicles is compatible with my past experience and habitual use of delivery service Compatibility (COA) Yoo et al. [14] COA2: Using autonomous delivery vehicles is comparable to the pickup distance and delivery price of my previous experience SI1: I can be infuenced by my peers’ preferences or decisions in choosing whether to use autonomous delivery vehicles Social infuence (SI) Hamad et al. [63] SI2: Te more people who have used autonomous delivery vehicles, the more willingness to use I have PR1: Te package the vehicle is carrying might be damaged or stolen PR2: Te delivery process may take too long PR3: Autonomous delivery vehicles will cause disclosure of private information such as address, name, and telephone number Perceived risk (PR) Yoo et al. [14]; Mathew et al. [1] PR4: Te remote-control center and the operating system of autonomous delivery vehicles may break down PR5: Hackers may attack the system and steal personal data, even operate autonomous delivery vehicles to attack pedestrians GI1: Using autonomous delivery vehicle services can reduce environmental pollution Green image (GI) Mathew et al. [1]; Li et al. [22] GI2: By using autonomous delivery vehicle services, I can demonstrate that I care about environmental conservation Infection risk (IR) IR: Using autonomous delivery vehicles can reduce the infection risk Tis study SA1: When communicating with strangers, I tend to feel uncomfortable Social awkwardness (SA) SA2: Using an autonomous delivery vehicle can remove the need for Tis study communication with couriers Use experience (UE) UE: Have you used autonomous delivery vehicles before? Tis study AI1: I Intend to use autonomous delivery vehicles as a delivery option in the future Adoption intention (AI) Kapser and Abdelrahman [16]; Hwang et al. [44] AI2: I am willing to pay more for autonomous delivery vehicle services 10 Journal of Advanced Transportation included in this survey. Responses containing what are 44.86% of them. Respondents with only a high school degree functionally contradictory “strongly agree” answers to such also account for the largest portion of respondents. Te questions, along with responses that were completed in less demographic analysis reveals that the sample is also rep- than 50 seconds, are excluded from analysis. resentative for the Chinese population in terms of age and gender [78]. Terefore, it is possible to generalize the fndings to a larger population in China. Te number of 3.3. Data Analysis. Our analysis includes two approaches, respondents who have used instant delivery service and namely, demographic analysis and structural equation those who have not (50.65% and 49.35% respectively) are modeling (SEM). Te demographic analysis includes a basic almost half and half. Takeaway food delivery is used most, profle of customers and the degree of infection risk for the followed by fresh food delivery and medicine delivery. It is location under study. Tese demographic characteristics worth noting that 56 respondents (8.1% of the survey such as age, gender, income, and education are analyzed as sample) indicated that they already have experience using control variables to capture their efects on customers’ ADV services. At the same time, as shown in Figure 2, adoption intentions. Te demographic analysis is followed 58.32% respondents are not willing to pay extra for ADV by SEM, which can be used to study the relationships be- service. Finally, the results shown in Figure 3 indicate that tween observed variables and latent variables [74]. SEM 58.93% of respondents remain neutral toward ADVs, which includes two stages-measurement model assessment and is consistent with previous research [16]. structural model assessment. Te measurement model assessment is performed to test 4.2. Measurement Model Assessment. Both the Cronbach’s the reliability and validity of the questionnaire. To validate alpha and CR for all the items of this study are found to be the model construct, four steps are implemented. Te frst above 0.70, except the Cronbach’s alpha for AI (Table 5). step is to test the internal construct reliability with Cron- According to a study by McCrae et al. [79], the reason for bach’s alpha and the composite reliability (CR), the outputs this exception may be the low number of questions re- of which should have thresholds higher than 0.70 [75, 76]. garding AI in the survey [80]. Te AVE values range from Second, convergent and discriminant validity should both be 0.612 to 0.926, which are all higher than 0.5, thus establishing examined. Convergent validity is assessed in terms of av- validity. Factor loadings range from 0.669 to 0.972 are all erage variance extracted (AVE) with a threshold of 0.50. In also higher than 0.5 [76] (Table 5). Te results show that addition, factor loadings should be higher than 0.5 [76]. convergent validity is established, indicating good agree- Tird, discriminant validity should be examined to measure ment between two or more items that measure the same how a construct is diferent from other constructs in the structure. Although the Cronbach’s alpha for AI is below SEM model by calculating the degree to which it correlates 0.70, the rest of parameters of AI including factor loadings, with other constructs and how distinctly it exists as a unique CR, and AVE all meet the standard, thus subsequent analysis construct [76]. In order to verify good discriminative val- can be conducted. √�� idity, the square root of AVE should be greater than the Te value of AVE is presented on the diagonal in correlation coefcient between the construct and other Table 6 and correlation coefcients are presented below the √�� constructs [76]. Finally, confrmatory factor analysis is used diagonal. Each variable has a greater AVE than the cor- to assess the measurement model [73]. Te parameters in relation coefcient, which demonstrates discriminant val- confrmatory factor analysis and their standards are drawn idity, i.e., that each construct is distinct from each other. from Cheung et al. [77]. Te results show that the proposed model has a greater Te structural model assessment is employed for path ft to the data based on the standards showed in Table 7. analysis. Te results include path coefcients and p-values. Only when the p-value <0.05 is the construct deemed to have a signifcant infuence on intention, thus supporting the 4.3. Structural Model Assessment. Te measurement model corresponding hypothesis. AMOS version 24 (Asymptoti- assessment is followed by structural model assessment cally distribution-free) is used for data analysis. [73, 76]. Te results, including path coefcients and p -values, are shown in Table 8 and Figure 4. Only when the p -value <0.05 is when the construct deemed to have a sig- 4. Results and Discussion nifcant infuence on intention, thus supporting the corre- sponding hypothesis. Te analysis from Table 8 shows that 4.1. Demographic Analysis. Te demographic details of re- spondents are shown in Table 4. Te percentages of re- eight of nine hypotheses are supported. More specifcally, compatibility (β � 0.171, p< 0.001), social infuence (β spondents who are male and female are 42.40% and 54.85%, respectively. Te percentage of respondents aged between � 0.205, p< 0.001), infection risk (β � 0.076, p< 0.05), green 25–34 and 35–49 years are 23.44% and 36.61%, respectively. image (β � 0.229, p< 0.001), social awkwardness (β � 0.302, Te distribution of other age groups generally mirrors that of p< 0.001), and experience (β � 0.040, p< 0.05) have a posi- the population studied, whereas the percentage of elderly tive infuence. Complexity (β � −0.294, p< 0.001) and per- over 65 is rather low due to the lower participation of the ceived risk (β � −0.149, p< 0.001) have a negative infuence elderly in using intelligent devices. Respondents with av- on customers’ intentions. Hence, hypotheses 2, 3, 4, 5, 6, 7, 8, and 9 are supported. However, contrary to expectation, erage monthly disposable incomes of 2000–6000 RMB make up the largest proportion of respondents, accounting for relative advantage (β � −0.058, p � 0.195) does not have Journal of Advanced Transportation 11 Table 4: Demographic details of respondents. Variable Category Frequency Percentage (%) Female 379 54.85 Gender Male 293 42.40 Tird gender 19 2.75 Below 18 years 115 16.64 19–24 96 13.89 25–34 162 23.44 Age 35–49 235 36.61 50–64 75 10.85 Above 65 years 8 1.16 Below 2000 RMB 143 20.69 2000–6000 RMB 310 44.86 6000–10000 RMB 120 17.37 Monthly household disposable income 10000–15000 RMB 58 8.39 15000–20000 RMB 21 3.04 Above 20000 RMB 39 5.64 Secondary school certifcate or below 28 4.05 High school degree 241 34.88 Bachelor’s degree 133 19.25 Education Master’s degree 213 30.82 Doctorate 58 8.39 No degree 18 2.60 Yes 350 50.65 Instant delivery use experience No 341 49.35 Fresh food 155 22.43 Take-away food 285 41.24 Instant delivery services used Medicine 89 12.88 Other 42 6.08 Yes 56 8.10 ADV use experience No 635 91.90 Medium and high-risk area 23 3.33 Epidemic risk level Low-risk area 668 96.67 a signifcant positive infuence. Te practical reason may be processes supporting the pickup and return of goods, which that the relative advantage in this context relates to whether are major components of complexity in the context of ADV ADVs can complete the delivery process and meet the de- service, should be as clear and efcient as possible to maximize customer ease of use of the service. An example livery demands of customers, while how goods are delivered is not the main concern of customers. Tis view was also process would be “face swiping,” which has become widely applied in check-in and -out services in Chinese railway expressed by many respondents during the ofine investigation. stations for picking up and returning goods. In addition to the variables related to our hypotheses, the In line with fndings from Yoo et al. [14], Min et al. [60], infuence of control variables (i.e., age, gender, income, and and Hamad et al. [63], our results show that compatibility education) on adoption intention is also studied (Table 9). has a positive efect on customers adoption intentions to- Except for gender, the control variables that were included in ward ADV services. Tis indicates that when ADV service is our survey are found to have signifcant efects on customers’ more consistent with ’customers’ previous delivery experi- intentions, although the efects sometimes diverge in direction. ence and habits, ’customers’ adoption intentions are higher. In addition, in comparison with the previous results without For example, traditional courier service enables goods to be delivered directly to one’s doorstep, but ADVs may not be control variables, the inclusion of age, gender, and income does not change the signifcance of any of the constructs (see Table 8). allowed to enter a given neighborhood or are not able to Tus, it is demonstrated that the model has good robustness. climb stairs. In these conditions, customers need to go Te results of path analysis provide evidence supporting downstairs to pick up goods. Terefore, some management DOI in the context of ADVs. Te results also indicate that regulations may need to be established to allow ADVs to infection risk, use experience, and social awkwardness are operate in particular residential areas, and ADVs may need important constructs afecting customer adoption intentions. to be equipped with RFID technology to communicate with the lift or parcel locker. Tus, home delivery can be ofered 4.4. Discussion. In accordance with fndings from previous and the convenience of using ADVs is improved. Kapser and studies [14, 60], complexity has a negative efect on cus- Abdelrahman [16] found that when the price is higher than tomers’ intention to adopt ADVs in instant delivery. Te the current price of courier delivery, customers are much less 12 Journal of Advanced Transportation Table 5: Results for factor loading, Cronbach’s alpha, CR, AVE, Extra cost respondents could accept and Sqrt of AVE. 1.74% 6.80% √�� Factor Cronbach’s Variable Item CR AVE AVE loading alpha RA1 0.905 RA2 0.905 RA 0.899 0.938 0.790 0.889 RA3 0.897 38.49% 52.97% RA4 0.848 COL1 0.919 COL COL2 0.957 0.928 0.959 0.885 0.941 COL3 0.946 COA1 0.958 COA 0.933 0.961 0.926 0.962 COA2 0.966 SI1 0.880 SI 0.724 0.905 0.827 0.909 SI2 0.938 No 10-20% PR1 0.916 <10% >20% PR2 0.918 Figure 2: Percentage of respondents according to acceptable in- PR PR3 0.900 0.905 0.969 0.863 0.929 crease in cost for ADV service. PR4 0.959 PR5 0.950 GI1 0.949 Attitude towards ADVs GI 0.934 0.960 0.923 0.961 GI2 0.972 4.49% SA1 0.951 SA 0.870 0.960 0.923 0.961 SA2 0.970 37.19% AI1 0.881 AI 0.573 0.759 0.612 0.782 AI2 0.669 Notes. RA � relative advantage; COL � complexity; COA � compatibility; 58.32% SI � social infuence; PR � perceived risk; GI � green image; SA � social awkwardness; AI � adoption intention. Table 6: Results for discriminant validity (Fornell-Larcker criterion). Positive RA COL COA SI PR GI SA Neutral RA 0.889 Negative COL −0.830 0.941 COA 0.756 −0.816 0.962 Figure 3: Percentage of respondents according to attitude towards SI 0.631 −0.698 0.672 0.909 ADVs. PR 0.321 −0.334 0.316 0.474 0.929 GI 0.584 −0.651 0.656 0.675 0.348 0.961 likely to accept ADVs. To increase customer use intentions, SA 0.340 −0.363 0.388 0.394 0.536 0.424 0.961 it is recommended that the price be equal to or lower than Notes. RA � relative advantage; COL � complexity; COA � compatibility; courier delivery. Tese fndings are also consistent with our SI � social infuence; PR � perceived risk; GI � green image; SA � social awkwardness. results as shown in Figure 2, with more than 90% of re- spondents indicating that they are not willing to pay more than 10% for ADV service compared to courier delivery. Table 7: Results of confrmatory factor analysis. Moreover, the ADVs should be available for service late at night and during severe weather like couriers. Te ADVs CMIN/ Indices χ df RMSEA NFI GFI AGFI should have the ability to operate in such settings, such as DF being equipped with specifc sensors and waterproof casing. Standards — — <3 <0.08 >0.7 >0.9 >0.7 Te results also confrm a positive efect of social network in Results 567.486 253 2.243 0.037 0.825 0.934 0.893 the context of ADVs, congruent with the fndings of Higgins et al. [67] and Min et al. [60]. Once one person starts using ADVs, more and more of their friends and family also begin to as issuing discount coupons for ADV services that can be shared with their friends to encourage them to try ADV services. We use ADVs, with such use difusing along their social networks. Terefore, the infuence of social networks can be utilized to suggest that ADV service providers invite individuals who have strong social infuence to try ADV services, similar to the efect promote the expansion of ADV service. For example, ADV service providers can introduce some incentive strategies such of celebrity product endorsements in advertisements. Journal of Advanced Transportation 13 Table 8: Results of path analysis. Hypothesis Path Proposed efect β Signifcance Result H1 RA⟶ AI + −0.058 0.195 Rejected H2 COL⟶ AI − −0.294 <0.001 Supported H3 COA⟶ AI + 0.171 <0.001 Supported H4 SI⟶ AI + 0.205 <0.001 Supported H5 PR⟶ AI − −0.149 <0.001 Supported H6 GI⟶ AI + 0.229 <0.001 Supported H7 IR⟶ AI + 0.076 <0.05 Supported H8 SA⟶ AI + 0.302 <0.001 Supported H9 UE⟶ AI + 0.040 <0.05 Supported Notes. RA � relative advantage; COL � complexity; COA � compatibility; SI � social infuence; PR � perceived risk; GI � green image; IR � infection risk; SA � social awkwardness; UE � use experience; AI � adoption intention. Relative advantage Complexity Compatibility 0.171* -0.294* -0.058 Social influence Infection risk 0.076* 0.205* Customer 0.302* -0.149* adoption Perceived risk Social awkwardness intention 0.040* 0.229* Green image Use experience 0.054* -0.014 0.058* -0.058* Age Gender Income Education Constructs from DOI Added Constructs Control variables Figure 4: Results of path analysis. Notes: means that this attribute is a signifcant contributor according to the analysis. Table 9: Control variable results. surrounding environment, so that the customer can track the ADVs in real-time via an application. Second, service Path β Signifcance providers should strengthen the robustness of their re- Age⟶ AI −0.058 <0.001 spective systems. For example, after an attack perpetrated by Gender⟶ AI −0.014 0.475 hackers, a backup system prepared in advance could be used Income⟶ AI 0.058 <0.01 Education⟶ AI 0.054 <0.05 to ensure uninterrupted service. Tird, with reference to the (H.R.3711) SELF DRIVE Act of America [81], ADV man- ufactures should establish a clear and reasonable data Along with the other positive fndings of this study, protection plan that encompasses the collection, use, a negative efect from perceived risk was also found, in line sharing, and storage of information about vehicle owners or with previous studies [1, 14]. In our study, perceived risk, customers, in order to enhance customer confdence in data including delivery, privacy, and technical risk, are found to protection. exert a negative efect on customer adoption intentions. Due Infection risk was rarely considered in previous studies. In to the need to upload a large amount of data such as ad- the present study, it has proven to be an important positive dresses online to successfully provide ADV services, cus- predictor of customer intentions. Contactless delivery has tomers are sensitive about protecting their privacy. Because become an efective method for reducing the spread of COVID-19 and other diseases while meeting daily needs. of the usage of the Internet in the process of delivery, system safety and robustness are also important components that ADVs play an important role in epidemic prevention and control. We suggest that more ADVs be deployed in medium- should be considered to ensure the smooth operation of the whole process and prevent hacking. To address these con- high infection risk areas, including hospital settings. cerns, we recommend the following three measures: First, Green image also demonstrated a signifcant positive the ADV should be installed with high-precision mapping efect on customer intentions, in agreement with previous and detection equipment to improve awareness of the studies [1, 14, 18]. Tese fndings indicate that protecting the 14 Journal of Advanced Transportation could be ofered to promote its use among those who can environment has become increasingly important to cus- tomers. ADVs may be an especially attractive option given serve as early adopters. that they are powered by electricity and have a lower adverse impact on the urban environment as compared to the use of 5. Conclusion delivery trucks and motorcycles. We suggest that the adoption of ADVs could be accelerated by promoting their In this study, we propose a customer adoption model that superior environmental performance as compared to the applies a DOI theory model with three new constructs and traditional option. Nonetheless, it should be acknowledged that focuses on customer adoption intention toward ADV that vehicle electrifcation and the carbon intensity of services in instant delivery. To our knowledge, this is the frst electricity still have key infuences on environmental study that considers impacts of use experience, infection performances [82]. risk, and social awkwardness in the context of ADVs. Eight Te results of the present study also indicate that social of nine constructs (i.e., complexity, compatibility, social awkwardness is an important positive predictor of cus- infuence, perceived risk, infection risk, green image, social tomers’ intentions. Perhaps due to the development of awkwardness, and use experience) are found to be statis- online shopping, social activities are decreasing, and people tically signifcant contributors to customer intentions to- are more accustomed to being alone or just communicating ward adopting ADV services. Furthermore, 58.32% of the with acquaintances. Direct contact with couriers may make respondents surveyed in this study expressed a neutral at- some customers feel awkward or even be perceived as po- titude toward ADV adoption, while 47.03% of respondents tentially dangerous, especially for women who live alone, expressed a willingness to pay more for ADV services thus making interactions with ADVs an acceptable or compared to courier delivery. Tis study also examines the a desirable alternative. Although social awkwardness pro- efects of age, gender, income, and education on adoption motes the use of ADVs, it may also aggravate the degree of intentions. Income and education are found to have sig- social isolation. To address this concern, ADV developers nifcant positive efects on the intention to adopt ADVs, can develop more humanized interactions between ADVs while age is found to serve as a negative predictor. Tese and humans during delivery services. results shed light on which socio-demographic factors exert Use experience is also found to positively afect customer the most infuence on adoption intentions towards ADV adoption intentions. Te results of path analysis confrm that service. customers who have already used ADVs are more likely to Overall, this study ofers some important insights into accept ADV services. Te reason may be that based on their the infuences of various constructs on customers’ intentions use experiences, their understanding of ADV services is to adopt ADVs in instant delivery. Tis study not only deeper and trust and confdence in ADVs services are in- enriches the research by providing a more refned theoretical creased by the direct use experience. Based on their previous framework for characterizing customer intentions toward experience with ADVs, they are more inclined to choose adopting ADV services, but also contributes to promoting ADVs as a delivery option. If more customers are provided the expansion of ADVs in instant delivery. It is the frst study use opportunities, for example through trial activities held in to investigate customer adoption intention toward ADVs some delimited areas such as industrial parks, then large- using DOI theory and broadening its application, rather scale use of ADV services could be achieved. than utilizing TAM theory, in recognition of the lack of end Lastly, age, income, and education are all found to be consumer concern for matters related to the direct use of signifcant contributors to customers’ adoption intentions. ADVs, with personal innovativeness and social infuence While age exhibited a negative efect on adoption intentions, being more prominent factors afecting customer intentions income and education both displayed positive efects. toward ADVs. New constructs are incorporated into the Compared to younger customers, older adults are more proposed model to adapt to the specifc context of ADVs, likely to be conservative and have less propensity to use new including infection risk, social awkwardness, and use ex- technologies. Tus, they are less willing to use driverless perience, with each found to be important predictors of vehicles for delivery. For customers with lower incomes, customer intention. In terms of practical implications, the a key reason that may discourage them from choosing ADVs fndings of this study can contribute to the formulation of is a concern about the delivery price. Tey may think that the relevant regulations (data privacy, security risks, etc.), delivery price of a new technology may be higher than the marketing strategies (such as discount coupons) and tech- traditional courier delivery. Tus, they are more likely to nology development (such as RFID and waterproofng) for choose courier delivery. Finally, customers with higher ADVs, furthering the popularization of their use. education may know more about ADVs. Terefore, they Te study is not without limitations. Te samples in- trust this new technology more and are more willing to volved in this study are all from China, so future research adopt ADVs. Based on the previous analysis, two sugges- should examine sample data covering diferent countries in tions are proposed as follows: (1) the use of ADVs could be order to validate the model proposed in this study and to promoted among young and/or more highly educated diferentiate geographical efects. In addition, only 8.1% of customers, so they can act as early adopters in their social respondents in our survey sample indicated prior ADV networks; with the assistance of such customers, older adults service use experience. Future research could investigate could gain more familiarity and comfort with ADV service customers with more experience with ADV service and processes; (2) a discount on the delivery price of the ADVs conduct multigroup analysis to compare adoption Journal of Advanced Transportation 15 [11] A. Pani, S. Mishra, M. Golias, and M. Figliozzi, “Evaluating intentions between non-users and users, identifying possible public acceptance of autonomous delivery robots during diferences in key variables such as the willingness to pay COVID-19 pandemic,” Transportation Research Part D: more. Another limitation is that it is a cross-sectional study Transport and Environment, vol. 89, 2020. that was conducted during the COVID-19 pandemic. [12] Udelv, “Autonomous delivery vehicles- why they matter, and Terefore, future investigation could include a longer in- how they work,” 2018. vestigation period and compare the results to postpandemic [13] Pengpai, “Meituan autonomous delivery vehicle crashed into periods. a private car, which was found to be entirely to blame: the ‘motor vehicle’ illegally entered a non-motorized lane,” 2021, Data Availability https://www.thepaper.cn/newsDetail_forward_14881840. [14] W. Yoo, E. Yu, and J. Jung, “Drone delivery: factors afecting Te sample data used to support the fndings of this study are the public’s attitude and intention to adopt,” Telematics and available from the corresponding author upon request. Informatics, vol. 35, no. 6, pp. 1687–1700, 2018. [15] Autocrypt, “When last Mile delivery turns autonomous – what are the considerations?,” 2021, https://autocrypt.io/ Conflicts of Interest when-last-mile-delivery-turns-autonomous/. [16] S. Kapser and M. Abdelrahman, “Acceptance of autonomous Te authors declare that there are no conficts of interest delivery vehicles for last-mile delivery in Germany – regarding the publication of this article. extending UTAUT2 with risk perceptions,” Transportation Research Part C: Emerging Technologies, vol. 111, pp. 210–225, Acknowledgments [17] B. V. Meldert and L. D. Boeck, “Introducing autonomous Tis work was supported in part by the National Natural vehicles in logistics a review from a broad perspective,” Report Science Foundation of China (72101188), the Shanghai KBI_1618, KU Leuven - Faculty of Economics and Business, Pujiang Program (2020PJC112), the Shanghai Municipal Leuven, Belgium, 2016. Science and Technology Major Project (2021SHZDZX0100), [18] J. Wu, H. Liao, J. W. Wang, and T. Chen, “Te role of en- and the Fundamental Research Funds for the Central vironmental concern in the public acceptance of autonomous Universities. Te authors would like to thank the editor and electric vehicles: a survey from China,” Transportation Re- reviewers for their insightful comments in improving the search Part F: Trafc Psychology and Behaviour, vol. 60, pp. 37–46, 2019. quality of this paper. [19] M. Liu, J. Wu, C. Zhu, and K. Hu, “Factors infuencing the acceptance of robo-taxi services in China: an extended References technology acceptance model analysis,” Journal of Advanced Transportation, vol. 2022, Article ID 8461212, 11 pages, 2022. [1] A. O. Mathew, A. N. Jha, A. K. Lingappa, and P. Sinha, [20] T. Dirsehan and C. Can, “Examination of trust and sus- “Attitude towards drone food delivery services—role of in- tainability concerns in autonomous vehicle adoption,” novativeness, perceived risk, and green image,” Journal of Technology in Society, vol. 63, Article ID 101361, 2020. Open Innovation: Technology, Market, and Complexity, vol. 7, [21] W. M. Chan and J. W. C. Lee, “5G connected autonomous no. 2, 2021. vehicle acceptance: the mediating efect of trust in the tech- [2] Foresight Research Institute, Analysis on the Market Status nology acceptance model,” Asian Journal of Business Research, Quo and Development Trend of China Instant Delivery In- vol. 11, no. 1, 2021. dustry in 2020, 2020. [22] H. Li, S. Yu, and J. Zheng, “Acceptance factors for younger [3] M. Lu, R. Wang, and P. Li, “Comparative analysis of online passengers in shared autonomous vehicles,” in Proceedings of fresh food shopping behavior during normal and COVID-19 the International Conference on Human-Computer In- crisis periods,” British Food Journal, vol. 124, no. 3, teraction, Virtual Event, July 2021. pp. 968–986, 2021. [23] S. Leon, C. Chen, and A. Ratclife, “Consumers’ perceptions of [4] P. Wang, X. Ma, and L. Hu, “Food delivery people become last mile drone delivery,” International Journal of Logistics a high incidence of trafc accidents,” 2020, http://www. Research and Applications, pp. 1–20, 2021. xinhuanet.com/mrdx/2020-09/15/c_139368949.htm. [24] M. Michels, C. F. von Hobe, P. J. Weller von Ahlefeld, and [5] AliResearch and Cainiao, National Socialized E-Commerce O. Musshof, “Te adoption of drones in German agriculture: Logistics Personnel Research Report, Jiaotong University, a structural equation model,” Precision Agriculture, vol. 22, Beijing, China, 2016. no. 6, pp. 1728–1748, 2021. [6] S. Chhabra, “Cost of last mile delivery for your business with [25] J. Y. Choe, J. J. Kim, and J. Hwang, “Innovative marketing ways to optimise it,” 2021, https://jungleworks.com/cost-of- strategies for the successful construction of drone food de- last-mile-delivery/. livery services: merging TAM with TPB,” Journal of Travel & [7] M. A. Figliozzi and D. Jennings, “A study of the competi- tiveness of autonomous delivery vehicles in urban areas,” Civil Tourism Marketing, vol. 38, no. 1, pp. 16–30, 2021. [26] A. M. H. Abrams, P. S. C. Dautzenberg, and C. Jakobowsky, and Environmental Engineering Faculty Publications and Presentations, 2020. “A theoretical and empirical refection on technology ac- ceptance models for autonomous delivery robots,” in Pro- [8] 36Kr, “Research report on autonomous delivery feld,” 2018, https://www.36kr.com/p/1725163094017. ceedings of the 2021 ACM/IEEE International Conference on Human-Robot Interaction, pp. 272–280, Boulder, CO, USA, [9] Flsenate, “CS/CS/HB 1289: autonomous vehicles,” 2021, https://fsenate.gov/Session/Bill/2021/1289. March 2021. [27] N. Adnan, S. Md Nordin, M. A. bin Bahruddin, and M. Ali, [10] Flsenate, “CS/CS/SB 1620: autonomous vehicles,” 2021, https://www.fsenate.gov/Session/Bill/2021/1620. “How trust can drive forward the user acceptance to the 16 Journal of Advanced Transportation technology? In-vehicle technology for autonomous vehicle,” [43] J. Berrada, I. Mouhoubi, and Z. Christoforou, “Factors of Transportation Research Part A: Policy and Practice, vol. 118, successful implementation and difusion of services based on pp. 819–836, 2018. autonomous vehicles: users’ acceptance and operators’ [28] C. Hewitt, I. Politis, and T. Amanatidis, “Assessing public proftability,” Research in Transportation Economics, vol. 83, perception of self-driving cars,” in Proceedings of the 24th Article ID 100902, 2020. International Conference on Intelligent User Interfaces, [44] J. Hwang, J. J. Kim, and K. W. Lee, “Investigating consumer pp. 518–527, Marina del Ray, CA, USA, March 2019. innovativeness in the context of drone food delivery services: [29] L. Meyer-Waaeden and J. Cloarec, “‘Baby, you can drive my its impact on attitude and behavioral intentions,” Techno- car’: psychological antecedents that drive consumers’ adop- logical Forecasting and Social Change, vol. 163, Article ID tion of AI-powered autonomous vehicles,” Technovation, 120433, 2021. vol. 109, 2021. [45] B. Aydin, “Public acceptance of drones: knowledge, attitudes, [30] S. Zhang, P. Jing, and G. Xu, “Te acceptance of independent and practice,” Technology in Society, vol. 59, 2019. autonomous vehicles and cooperative vehicle-highway au- [46] L. K. Lin Tan, B. C. Lim, and G. Park, “Public acceptance of tonomous vehicles,” Information, vol. 12, no. 9, 2021. drone applications in a highly urbanized environment,” [31] K. B. Kim and B. G. Chung, “Technology acceptance of in- Technology in Society, vol. 64, 2021. dustry 4.0 applying UTAUT2: focusing on AR ang drone [47] X. Zhu, T. J. Pasch, and A. Bergstrom, “Understanding the services,” Journal of Information Technology, vol. 26, no. 6, structure of risk belief systems concerning drone delivery: pp. 29–46, 2019. a network analysis,” Technology in Society, vol. 62, Article ID [32] H. Ganjipour and A. Edrisi, “Applying the integrated model to 101262, 2020. understanding online buyers’ intention to adopt delivery [48] M. Yarwood, Psychology of Human Emotion. Afordable drones in Iran,” Transportation Letters, vol. 15, no. 2, Course Transformation, Pennsylvania State University, pp. 98–110, 2022. Pennsylvania, PA, USA, 2022. [33] S. Kapser, M. Abdelrahman, and T. Bernecker, “Autonomous [49] W. Duan and G. Jiang, “A review of the theory of planned delivery vehicles to fght the spread of Covid-19 - how do men behavior,” Advances in Psychological Science, vol. 16, no. 2, and women difer in their acceptance?” Transportation Re- pp. 315–320, 2008. search Part A: Policy and Practice, vol. 148, pp. 183–198, 2021. [50] W. Qazi, S. A. Raza, and N. Shah, “Acceptance of e-book [34] M. A. Ribeiro, D. Gursoy, and O. H. Chi, “Customer ac- reading among higher education students in a developing ceptance of autonomous vehicles in travel and tourism,” country the modifed difusion innovation theory,” In- Journal of Travel Research, vol. 61, no. 3, pp. 620–636, 2021. ternational Journal of Business Information Systems, vol. 27, [35] Z. B. Ramadan, M. F. Farah, and M. Mrad, “An adapted TPB no. 2, pp. 222–245, 2018. approach to consumers’ acceptance of service-delivery [51] E. M. Rogers, “Difusion of innovation theory,” 1995, https:// drones,” Technology Analysis & Strategic Management, sphweb.bumc.bu.edu/otlt/mph-modules/sb/ vol. 29, no. 7, pp. 817–828, 2016. behavioralchangetheories/behavioralchangetheories4.html. [36] P. Liu, Q. Guo, F. Ren, L. Wang, and Z. Xu, “Willingness to [52] E. M. Rogers, Difusion or Innovations, Free Press, New York, pay for self-driving vehicles: infuences of demographic and NY, USA, 2003. psychological factors,” Transportation Research Part C: [53] Z. Rezvani, J. Jansson, and J. Bodin, “Advances in consumer Emerging Technologies, vol. 100, pp. 306–317, 2019. electric vehicle adoption research: a review and research [37] M. L. Cunningham, M. A. Regan, S. A. Ledger, and agenda,” Transportation Research Part D: Transport and J. M. Bennett, “To buy or not to buy? Predicting willingness to Environment, vol. 34, pp. 122–136, 2015. pay for automated vehicles based on public opinion,” [54] R. Agarwal and J. Prasad, “Te role of innovation charac- Transportation Research Part F: Trafc Psychology and Be- teristics and perceived voluntariness in the acceptance of haviour, vol. 65, pp. 418–438, 2019. information technologies,” Decision Sciences, vol. 28, no. 3, [38] L. M. Hulse, H. Xie, and E. R. Galea, “Perceptions of au- pp. 557–582, 1997. tonomous vehicles: relationships with road users, risk, gender [55] K. B. Ooi, J. J. Sim, K. T. Yew, and B. Lin, “Exploring factors and age,” Safety Science, vol. 102, pp. 1–13, 2018. infuencing consumers’ behavioral intention to adopt [39] R. Shabanpour, N. Golshani, A. Shamshiripour, and broadband in Malaysia,” Computers in Human Behavior, A. K. Mohammadian, “Eliciting preferences for adoption of vol. 27, no. 3, pp. 1168–1178, 2011. fully automated vehicles using best-worst analysis,” Trans- [56] Y. S. Wang, S. C. Wu, H. H. Lin, Y. M. Wang, and T. R. He, portation Research Part C: Emerging Technologies, vol. 93, “Determinants of user adoption of web ‘automatic teller pp. 463–478, 2018. machines’: an integrated model of ’transaction cost theory’ [40] C. A. Spurlock, J. Sears, G. Wong-Parodi et al., “Describing and ’innovation difusion theory,” Service Industries Journal, the users: understanding adoption of and interest in shared, vol. 32, no. 9, pp. 1505–1525, 2012. electrifed, and automated transportation in the San Francisco [57] K. C. Chung, “Gender, culture and determinants of behav- Bay Area,” Transportation Research Part D: Transport and ioural intents to adopt mobile commerce among the Y Environment, vol. 71, pp. 283–301, 2019. Generation in transition economies: evidence from [41] C. Bernhard, D. Oberfeld, C. Hofmann, D. Weismuller, and Kazakhstan,” Behaviour & Information Technology, vol. 33, H. Hecht, “User acceptance of automated public transport,” no. 7, pp. 743–756, 2013. Transportation Research Part F: Trafc Psychology and Be- [58] J. Y. M. Kang, J. M. Mun, and K. K. P. Johnson, “In-store haviour, vol. 70, pp. 109–123, 2020. mobile usage: downloading and usage intention toward [42] K. F. Yuen, L. Cai, G. Qi, and X. Wang, “Factors infuencing autonomous vehicle adoption: an application of the tech- mobile location-based retail apps,” Computers in Human nology acceptance model and innovation difusion theory,” Behavior, vol. 46, pp. 210–217, 2015. Technology Analysis & Strategic Management, vol. 33, no. 5, [59] M. S. Sharifzadeh, C. A. Damalas, and G. Abdollahzadeh, pp. 505–519, 2020. “Predicting adoption of biological control among Iranian rice Journal of Advanced Transportation 17 farmers: an application of the extended technology acceptance [78] National Bureau of Statistics, “Bulletin of the seventh national model (TAM2),” Crop Protection, vol. 96, pp. 88–96, 2017. census (4th),” 2021, http://www.stats.gov.cn/english/ PressRelease/202105/t20210510_1817185.html. [60] S. Min, K. K. F. So, and M. Jeong, “Consumer adoption of the [79] R. R. McCrae, J. E. Kurtz, S. Yamagata, and A. Terracciano, Uber mobile application: insights from difusion of in- “Internal consistency, retest reliability, and their implications novation theory and technology acceptance model,” Journal of for personality scale validity,” Personality and Social Psy- Travel & Tourism Marketing, vol. 36, no. 7, pp. 770–783, 2018. chology Review, vol. 15, no. 1, pp. 28–50, 2011. [61] S. Nordhof, J. de Winter, R. Madigan, N. Merat, B. van Arem, [80] M. Tavakol and R. Dennick, “Making sense of Cronbach’s and R. Happee, “User acceptance of automated shuttles in alpha,” International Journal of Medical Education, vol. 2, Berlin-Schoneberg: ¨ a questionnaire study,” Transportation pp. 53–55, 2011. Research Part F: Trafc Psychology and Behaviour, vol. 58, [81] Congress, “Privacy plan required for highly automated ve- no. 58, pp. 843–854, 2018. hicles,” 2021, https://www.congress.gov/bill/117th-congress/ [62] Y. Duan, Q. He, W. Feng, D. Li, and Z. Fu, “A study on e-learning house-bill/3711/text?r=1. take-up intention from an innovation adoption perspective: a case [82] L. Li, X. He, G. A. Keoleian et al., “Life cycle greenhouse gas in China,” Computers & Education, vol. 55, no. 1, pp. 237–246, emissions for last-mile parcel delivery by automated vehicles and robots,” Environmental Science & Technology, vol. 55, [63] A. A. A. Hamad, I. Petri, Y. Rezgui, and A. Kwan, “Towards no. 16, Article ID 11360, 2021. the innovation of an integrated ‘One-Stop-Shop’ online services utility management: exploring customer’ technology acceptance,” Sustainable Cities and Society, vol. 34, pp. 126– 143, 2017. [64] Rahul, “Autonomous vehicles deliveries vs. Drone deliveries,” 2019, https://blog.route4me.com/autonomous-vehicle- deliveries-vs-drone-deliveries/. [65] Tencent, “Shocking! A delivery man urinated in a customer’s food,” 2021, https://new.qq.com/omn/20211205/ 20211205A02NM000.html. [66] H. P. Young, “Innovation difusion in heterogeneous pop- ulations: contagion, social infuence, and social learning,” Te American Economic Review, vol. 99, no. 5, pp. 1899–1924, [67] C. Higgins, D. Compeau, and D. Meister, “From prediction to explanation: reconceptualizing and extending the perceived characteristics of innovating,” Journal of the Association for Information Systems, vol. 8, no. 8, pp. 409–439, 2007. [68] J. P. Roderigue, Te Geography of Transport System, Rout- ledge, New York, NY, USA, 2020. [69] M. A. Figliozzi, “Carbon emissions reductions in last mile and grocery deliveries utilizing air and ground autonomous ve- hicles,” Transportation Research Part D: Transport and En- vironment, vol. 85, 2020. [70] G. Xi, X. Cao, and F. Zhen, “How does same-day-delivery online shopping reshape social interactions among neighbors in Nanjing?” Cities, vol. 114, Article ID 103219, 2021. [71] X. Li, “White rhino comes into our life, whose delivery ca- pabilities are more than couriers,” Jiading Newspaper, 2021. [72] A. L. Comrey and H. B. Lee, A First Course in Factor Analysis, Erlbaum, Hillsdale, MI, USA, 1992. [73] R. B. Kline, Principles and Practice of Structural Equation Modeling, Guilford Publications, New York, NY, USA, 2016. [74] T. Zhang, D. Tao, X. Qu et al., “Automated vehicle acceptance in China: social infuence and initial trust are key de- terminants,” Transportation Research Part C: Emerging Technologies, vol. 112, pp. 220–233, 2020. [75] C. Fornell and D. F. Larcker, “Evaluating structural equation models with unobservable variables and measurement error,” Journal of Marketing Research, vol. 18, no. 1, pp. 39–50, 1981. [76] J. F. Hair, W. C. Black, and B. J. Babin, Multivariate Data Analysis, Cengage Learning, Boston, MA, USA, 2014. [77] C. M. Cheung, R. P. Zhang, R. Wang, S. C. Hsu, and P. Manu, “Group-level safety climate in the construction industry: infuence of organizational, group, and individual factors,” Journal of Management in Engineering, vol. 38, no. 1, 2022. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Advanced Transportation Hindawi Publishing Corporation http://www.deepdyve.com/lp/hindawi-publishing-corporation/customer-s-adoption-intentions-toward-autonomous-delivery-vehicle-hRXyGpryUc

Loading next page...

References (73)

Ramin Shabanpour, Nima Golshani, Ali Shamshiripour, A. Mohammadian (2018)
Eliciting preferences for adoption of fully automated vehicles using best-worst analysis
Transportation Research Part C: Emerging Technologies
Duan Wen (2008)
A Review of the Theory of Planned Behavior
(2018)
Research report on autonomous delivery field
Jingwen Wu, Hua Liao, Jin-Wei Wang, Tianqi Chen (2019)
The role of environmental concern in the public acceptance of autonomous electric vehicles: A survey from China
Transportation Research Part F: Traffic Psychology and Behaviour
Ki-bong Kim, Chung， Gyu (2019)
Technology Acceptance of Industry 4.0 Applying UTAUT2:Focusing on AR and Drone Services
, 26
C. Cheung, R. Zhang, Ran Wang, S. Hsu, P. Manu (2022)
Group-Level Safety Climate in the Construction Industry: Influence of Organizational, Group, and Individual Factors
Journal of Management in Engineering
C. Spurlock, James Sears, G. Wong‐Parodi, Victor Walker, Ling Jin, Margaret Taylor, Andrew Duvall, Anand Gopal, A. Todd (2019)
Describing the users: Understanding adoption of and interest in shared, electrified, and automated transportation in the San Francisco Bay Area
Transportation Research Part D: Transport and Environment
S. Nordhoff, J. Winter, R. Madigan, N. Merat, B. Arem, R. Happee (2018)
User acceptance of automated shuttles in Berlin-Schöneberg: A questionnaire study
Transportation Research Part F: Traffic Psychology and Behaviour
Z. Rezvani, Johan Jansson, J. Bodin (2015)
Advances in consumer electric vehicle adoption research: A review and research agenda
Transportation Research Part D-transport and Environment, 34
Tingru Zhang, Da Tao, Xingda Qu, Xiaoyan Zhang, Jihong Zeng, Haoyu Zhu, Han Zhu (2020)
Automated vehicle acceptance in China: Social influence and initial trust are key determinants
Transportation Research Part C-emerging Technologies, 112
(2021)
White rhino comes into our life, whose delivery capabilities are more than couriers
(2021)
Cost of last mile delivery for your business with ways to optimise it
(2021)
Shocking! A delivery man urinated in a customer's food
Taşkın Dirsehan, Ceren Can (2020)
Examination of trust and sustainability concerns in autonomous vehicle adoption
Technology in Society, 63
Shuang-xi Zhang, Peng Jing, Gang Xu (2021)
The Acceptance of Independent Autonomous Vehicles and Cooperative Vehicle-Highway Autonomous Vehicles
Inf., 12
R. McCrae, J. Kurtz, S. Yamagata, A. Terracciano (2011)
Internal Consistency, Retest Reliability, and Their Implications for Personality Scale Validity
Personality and Social Psychology Review, 15
C. Higgins, Deborah Compeau, D. Meister (2007)
From Prediction to Explanation: Reconceptualizing and Extending the Perceived Characteristics of Innovating
J. Assoc. Inf. Syst., 8
(2021)
Meituan autonomous delivery vehicle crashed into a private car, which was found to be entirely to blame: the ‘motor vehicle’ illegally entered a non-motorized lane
(2020)
Food delivery people become a high incidence of traffic accidents
Z. Ramadan, M. Farah, Mona Mrad (2017)
An adapted TPB approach to consumers’ acceptance of service-delivery drones
Technology Analysis & Strategic Management, 29
Wasim Qazi, S. Raza, Nida Shah (2018)
Acceptance of e-book reading among higher education students in a developing country: the modified diffusion innovation theory
Int. J. Bus. Inf. Syst., 27
M. Michels, Cord-Friedrich Hobe, Paul Ahlefeld, O. Musshoff (2021)
The adoption of drones in German agriculture: a structural equation model
Precision Agriculture, 22
J. Berrada, I. Mouhoubi, Z. Christoforou (2020)
Factors of successful implementation and diffusion of services based on autonomous vehicles: users’ acceptance and operators’ profitability
Research in Transportation Economics
A. Mathew, Abhishek Jha, Anasuya Lingappa, Pranshu Sinha (2021)
Attitude towards Drone Food Delivery Services—Role of Innovativeness, Perceived Risk, and Green Image
Journal of Open Innovation: Technology, Market, and Complexity, 7
Mingyu Liu, Jianping Wu, Chunli Zhu, Kezhen Hu (2022)
Factors Influencing the Acceptance of Robo-Taxi Services in China: An Extended Technology Acceptance Model Analysis
Journal of Advanced Transportation
Guangliang Xi, X. Cao, F. Zhen (2021)
How does same-day-delivery online shopping reshape social interactions among neighbors in Nanjing?
Cities
Charles Hewitt, I. Politis, Theocharis Amanatidis, Advait Sarkar (2019)
Assessing public perception of self-driving cars: the autonomous vehicle acceptance model
Proceedings of the 24th International Conference on Intelligent User Interfaces
Yi-Shun Wang, Shun-Cheng Wu, Hsin‐Hui Lin, Yu-Min Wang, T. He (2012)
Determinants of user adoption of web ''Automatic Teller Machines': an integrated model of 'Transaction Cost Theory' and 'Innovation Diffusion Theory'
The Service Industries Journal, 32
Luyao Li, Xiaoyi He, G. Keoleian, Hyung Kim, Robert Kleine, T. Wallington, Nicholas Kemp (2021)
Life Cycle Greenhouse Gas Emissions for Last-Mile Parcel Delivery by Automated Vehicles and Robots.
Environmental science & technology
Joshua Epstein, Edoardo Gallo, James Heckman, J. Hofbauer, C. Manski, David Myatt, Thomas Norman, T. Valente, D. Watts (2009)
Innovation Diffusion in Heterogeneous Populations: Contagion, Social Influence, and Social Learning
The American Economic Review, 99
(2021)
When last Mile delivery turns autonomous – what are the considerations?
M. Figliozzi, Dylan Jennings (2020)
A Study of the Competitiveness of Autonomous Delivery Vehicles in Urban Areas
M. Ribeiro, D. Gursoy, Oscar Chi (2021)
Customer Acceptance of Autonomous Vehicles in Travel and Tourism
Journal of Travel Research, 61
Sebastian Kapser, Mahmoud Abdelrahman, Tobias Bernecker (2021)
Autonomous delivery vehicles to fight the spread of Covid-19 – How do men and women differ in their acceptance?
Transportation Research. Part A, Policy and Practice, 148
Ritu Agarwal, J. Prasad (1997)
The Role of Innovation Characteristics and Perceived Voluntariness in the Acceptance of Information Technologies
Decision Sciences, 28
M. Sharifzadeh, C. Damalas, G. Abdollahzadeh, Hossein Ahmadi-Gorgi (2017)
Predicting adoption of biological control among Iranian rice farmers: An application of the extended technology acceptance model (TAM2)
Crop Protection, 96
K. Ooi, J. Sim, King-Tak Yew, Binshan Lin (2011)
Exploring factors influencing consumers' behavioral intention to adopt broadband in Malaysia
Comput. Hum. Behav., 27
C. Bernhard, D. Oberfeld, C. Hoffmann, Dirk Weismüller, H. Hecht (2020)
User acceptance of automated public transport
Transportation Research Part F-traffic Psychology and Behaviour, 70
L. Meyer-Waarden, Julien Cloarec (2021)
“Baby, you can drive my car”: Psychological antecedents that drive consumers’ adoption of AI-powered autonomous vehicles
Technovation
Wonsang Yoo, Eun-Ju Yu, Jaemin Jung (2018)
Drone delivery: Factors affecting the public's attitude and intention to adopt
Telematics Informatics, 35
Hao Li, Sisi Yu, Jiatai Zheng, Xue Zhao, Peifang Du, Hao Tan (2021)
Acceptance Factors for Younger Passengers in Shared Autonomous Vehicles
C. Fornell, D. Larcker (1981)
Evaluating structural equation models with unobservable variables and measurement error.
Journal of Marketing Research, 18
Miaojia Lu, Ran Wang, Peiyang Li (2021)
Comparative analysis of online fresh food shopping behavior during normal and COVID-19 crisis periods
British Food Journal
Steven Leon, Charlie Chen, Aaron Ratcliffe (2021)
Consumers’ perceptions of last mile drone delivery
International Journal of Logistics Research and Applications, 26
Kum Yuen, Lanhui Cai, G. Qi, Xueqin Wang (2020)
Factors influencing autonomous vehicle adoption: an application of the technology acceptance model and innovation diffusion theory
Technology Analysis & Strategic Management, 33
Alsulaiman Hamad, I. Petri, Y. Rezgui, A. Kwan (2017)
Towards the innovation of an integrated 'one-stop-shop' online services utility management: exploring customer' technology acceptance
Sustainable Cities and Society, 34
(2018)
Autonomous delivery vehicles- why they matter, and how they work
M. Hornor (2007)
Diffusion of Innovation Theory
Xun Zhu, T. Pasch, A. Bergstrom (2020)
Understanding the structure of risk belief systems concerning drone delivery: A network analysis
Technology in Society, 62
K. Chung (2014)
Gender, culture and determinants of behavioural intents to adopt mobile commerce among the Y Generation in transition economies: evidence from Kazakhstan
Behaviour & Information Technology, 33
Sebastian Kapser, Mahmoud Abdelrahman (2020)
Acceptance of autonomous delivery vehicles for last-mile delivery in Germany – Extending UTAUT2 with risk perceptions
Transportation Research Part C: Emerging Technologies
(2021)
Privacy plan required for highly automated vehicles
J. Choe, J. Kim, Jinsoo Hwang (2021)
Innovative marketing strategies for the successful construction of drone food delivery services: Merging TAM with TPB
Journal of Travel & Tourism Marketing, 38
Somang Min, Kevin So, Miyoung Jeong (2018)
Consumer adoption of the Uber mobile application: Insights from diffusion of innovation theory and technology acceptance model
Journal of Travel & Tourism Marketing, 36
N. Adnan, S. Nordin, Mohamad Bahruddin, Murad Ali (2018)
How trust can drive forward the user acceptance to the technology? In-vehicle technology for autonomous vehicle
Transportation Research Part A: Policy and Practice
W. Chan, J. Lee (2021)
5G Connected Autonomous Vehicle Acceptance: The Mediating Effect of Trust in the Technology Acceptance Model
, 11
(2021)
Bulletin of the seventh national census (4th)
(2020)
The Geography of Transport Systems
H. Ganjipour, A. Edrisi (2022)
Applying the integrated model to understanding online buyers’ intention to adopt delivery drones in Iran
Transportation Letters, 15
Anna Abrams, Pia Dautzenberg, Carla Jakobowsky, Stefan Ladwig, Astrid Pütten (2021)
A Theoretical and Empirical Reflection on Technology Acceptance Models for Autonomous Delivery Robots
2021 16th ACM/IEEE International Conference on Human-Robot Interaction (HRI)
Y. Duan, Qile He, W. Feng, Daoliang Li, Zetian Fu (2010)
A study on e-learning take-up intention from an innovation adoption perspective: A case in China
Comput. Educ., 55
(2019)
Autonomous vehicles deliveries vs. Drone deliveries
Jinsoo Hwang, J. Kim, Kwang-Woo Lee (2020)
Investigating consumer innovativeness in the context of drone food delivery services: Its impact on attitude and behavioral intentions
Technological Forecasting and Social Change
Ju-Young Kang, J. Mun, K. Johnson (2015)
In-store mobile usage: Downloading and usage intention toward mobile location-based retail apps
Comput. Hum. Behav., 46
L. Hulse, H. Xie, E. Galea (2018)
Perceptions of autonomous vehicles: Relationships with road users, risk, gender and age
Safety Science, 102
Lynn Tan, Beng-Chong Lim, Guihyun Park, K. Low, Victor Yeo (2021)
Public acceptance of drone applications in a highly urbanized environment
Technology in Society
Agnivesh Pani, Sabyasachee Mishra, M. Golias, M. Figliozzi (2020)
Evaluating public acceptance of autonomous delivery robots during COVID-19 pandemic
Transportation Research Part D: Transport and Environment
Peng Liu, Qianru Guo, Fei Ren, Lin Wang, Zhigang Xu (2019)
Willingness to pay for self-driving vehicles: Influences of demographic and psychological factors
Transportation Research Part C: Emerging Technologies
M. Tavakol, R. Dennick (2011)
Making sense of Cronbach's alpha
International Journal of Medical Education, 2
M. Figliozzi (2020)
Carbon emissions reductions in last mile and grocery deliveries utilizing air and ground autonomous vehicles
Transportation Research. Part D, Transport and Environment, 85
(2021)
CS/CS/HB 1289: autonomous vehicles
Burchan Aydin (2019)
Public acceptance of drones: Knowledge, attitudes, and practice
Technology in Society
M. Cunningham, M. Regan, Selena Ledger, Joanne Bennett (2019)
To buy or not to buy? Predicting willingness to pay for automated vehicles based on public opinion
Transportation Research Part F: Traffic Psychology and Behaviour

Publisher: Hindawi Publishing Corporation
ISSN: 0197-6729
eISSN: 2042-3195
DOI: 10.1155/2023/3440691
Publisher site: See Article on Publisher Site

Abstract

Hindawi Journal of Advanced Transportation Volume 2023, Article ID 3440691, 17 pages https://doi.org/10.1155/2023/3440691 Research Article Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience 1 1 2 1 Miaojia Lu , Chengyuan Huang , Ran Wang , and Hao Li College of Transportation Engineering, Tongji University, Shanghai 201804, China College of Civil Engineering, Hunan University, Changsha 410082, China Correspondence should be addressed to Chengyuan Huang; hcyuann@163.com Received 16 September 2022; Revised 5 February 2023; Accepted 17 February 2023; Published 8 March 2023 Academic Editor: Michela Le Pira Copyright © 2023 Miaojia Lu et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Te high demand and acute timeliness that characterizes instant delivery entail the challenges of high labor costs and an increase in courier trafc accidents. Autonomous delivery vehicles (ADVs) may serve as a key solution, with their attendant reduced labor input and higher efciency. Customers play a key role in the successful implementation of ADVs on a large scale. However, understanding the factors that afect customers’ intentions to use ADVs is still limited. Compared to autonomous driving, ADV customers are ultimately not the real users, who only are served by ADVs during the last leg of a trip. On account of this, the Technology Acceptance Model (TAM) may not be well-ftted for explaining the dynamics involved in ADV adoption. Within the context of ADVs, our study identifed infuencing factors that have not been captured by prior studies. Tis study incorporates infection risk, use experience, and social awkwardness into the Difusion of Innovation (DOI) theory to explore customers’ intentions to use ADVs. Data from 691 survey respondents were collected to validate the research design. Te results demonstrate that compatibility, social infuence, infection risk, green image, social awkwardness, and use experience all have a signifcantly positive impact on customers’ intentions to adopt ADV services, while complexity and perceived risk both exhibited a negative impact. But no efect could be found for relative advantage, which may be because of the fact that customers only need ADVs to meet their delivery demand. Tis study contributes to understanding customers’ adoption intentions toward ADVs, informing policymakers in formulating ADV regulations and standards, and promoting the large-scale application of ADVs in instant delivery services. accidents related to instant delivery. In August 2020, ap- 1. Introduction proximately 12 thousand trafc violations related to instant With the acceleration of the pace of life, combined with the delivery occurred in Shenzhen, China [4]. In addition, the ease and convenience of online shopping and increases in working-age population in China has been declining since income, more and more people have chosen to use instant 2013 and nearly 25% of couriers work more than 12 hours delivery services [1]. Instant delivery orders include not only a day [5]. Te concomitant rising human labor costs account meals but also fresh food, fowers, and other products such for 50 to 60% of the total cost of last-mile delivery [6]. as medicine and important documents. Te scale of instant Terefore, many startup companies and Internet companies delivery reached 24.37 billion orders in China in the year such as Amazon, Nuro, Jingdong, and Meituan [7] have 2020, a ten-fold increase since 2015 [2]. In particular, the begun to explore innovative delivery technologies such as frequency of customers shopping online for fresh food grew Autonomous Delivery Vehicles (ADVs), which are self- 71.2% during the COVID-19 pandemic [3]. Te increasing driving vehicles that travel on roads (as distinct from order volume, the importance of freshness, and the demand other autonomous robots that travel on sidewalks) to deliver for timeliness have consequently led to more courier trafc goods. ADVs can reduce labor costs and accident rates, as 2 Journal of Advanced Transportation intentions can aid in advancing large-scale application of well as increase delivery efciency, as one ADV can replace two couriers [8]. In the United States, Europe, and China, ADV technology. In the meantime, legislators are tackling how to properly regulate the use of ADVs. Te fndings of many regulations have been enacted to support the public road operation of ADVs, with corresponding operation this study as they relate to customer behaviors can also assist requirements [7, 9, 10]. Due to the often intensely chal- in the formulation of sound legislation. lenging conditions that have arisen from the COVID-19 Te remainder of the study is organized as follows: pandemic, customers, businesses, and governments have Section 2 reviews previous studies about the acceptance of switched from being cautious beta testers to becoming eager autonomous vehicles (AVs), ADVs, unmanned aerial ve- early adopters [11]. Te novelty, convenience, efciency, and hicles (UAVs), and autonomous robots, as well as studies using DOI theory. Section 3 details the constructs and contactless delivery associated with ADVs have led cus- tomers to develop high expectations for this technology [12]. hypotheses in the proposed theoretical model. Tis section also describes the questionnaire designed to investigate However, many customers are still skeptical and cau- tious about the viability of ADVs. Accidents involving au- potential customers’ intentions and the methods used to analyze the collected data. Section 4 presents the descriptive tonomous vehicles are worrisome to the public. In October 2021, a trafc accident involving an ADV and a private car statistics and discusses the results of the model assessments. took place in Beijing, with the ADV found to be primarily Section 5 recaps some key contributions of this study, in- responsible for the accident [13]. On August 14, 2021, an dicates limitations of the study, and suggests directions for entrepreneur died while driving NIO’s ES8 SUV when the future research. automatic cruise system was turned on. In addition, ADVs may break down on the way or lose customers’ goods [1, 14]. 2. Literature Review Moreover, the massive amount of data required for oper- ating autonomous vehicles has raised concerns about pri- 2.1. Factors and Teories Related to Autonomous Vehicles vacy. Some have expressed alarm about the risk of hackers Acceptance and Intentions. Here, we review previous studies infltrating ADV systems and controlling ADVs to attack related to acceptance and intentions toward autonomous pedestrians [15]. It has been argued that the biggest barrier vehicles, with particular attention to autonomous driving to widespread adoption of ADVs is not technological ca- and autonomous delivery. Teoretical models used in pre- pability but social acceptance. Customers play a key role in vious research are shown in Table 1. the successful implementation of ADVs on a large scale [16]. Terefore, factors that may have signifcant infuences on customers’ adoption intentions need to be explored to ex- 2.1.1. Autonomous Driving. Autonomous driving refers to pand ADV acceptance as well as to integrate their consid- cars capable of sensing its environment and operating eration in the development of company logistics and without human involvement. Many researchers have in- government regulations. Te aim of this study is to in- vestigated the acceptance of autonomous driving. Some vestigate factors that signifcantly infuence customers’ in- studies have been conducted based on behavioral theories tentions to use ADV, drawing upon behavior theories. [18, 21, 28, 34], while other studies have investigated par- Difusion of Innovation (DOI) theory is applied in this ticular factors to explain user acceptance [36–41]. study because of the following reasons: (1) it can be used to Perceived usefulness and perceived ease of use are both explore the characteristics of personal innovativeness and important factors proven to afect user acceptance and in- (2) this theory is suitable for examining the customers, who tention [18, 19, 42]. Many researchers add other factors such are not the real users of ADVs [17] but are served by ADVs as green image [18] and perceived trust [19, 22] into the during the last leg of a trip. Our study also identifes factors Technology Acceptance Model (TAM) for further analytical afecting adoption that have not been captured in previous refnement. Using the Unifed Teory of Acceptance and Use studies. ADVs have played an important role in the delivery of Technology (UTAUT), Adnan et al. [27], Hewitt et al. of daily necessities and medicine during the COVID-19 [28], and Meyer_Waaeden and Cloarec [29] all investigated pandemic because of their ability to facilitate contactless the importance of performance expectancy, efort expec- delivery. Given this key characteristic of ADVs and the tancy, trust, and social infuence in AV adoption. Liu et al. ongoing public health context in which ADVs operate, [36] studied customers’ willingness to pay for self-driving infection risk is considered in this study. Because ADVs can vehicles and they found that more than 50% of respondents bypass person-to-person contact, another factor, i.e., social were willing to pay extra, although the amounts are diferent. awkwardness, is also considered. Use experience of ADVs is In addition, Berrada et al. [43] divided customers into fve also incorporated into the research design. Tis study is the clusters including conservatives, skeptics, late adopters, early frst to analyze the adoption intention of customers with adopters, and explorers and found more than 60% of re- regard to ADVs based on DOI theory. Tis study also spondents were willing to use autonomous vehicles. Similar contributes to existing research by considering infection to this study, Bernhard et al. [41] conducted two subsurveys risk, use experience, and social awkwardness in relation to to investigate the efects of autonomous minibus using ADV adoption intentions. Gaining a more comprehensive experience and found that respondents who already had understanding of the elements that infuence adoption experience using autonomous minibuses were more willing Journal of Advanced Transportation 3 Table 1: Teory models applied in previous studies of autonomous driving and autonomous delivery. Teory Teory description References AVs: Wu et al. [18]; Liu et al. [19]; Dirsehan and Can [20]; Chan TAM is used to explain how perceived usefulness and perceived and Lee [21]; Li et al. [22] Technology acceptance model (TAM) ease of use afect personal behaviors [18] UAVs: Leon et al. [23]; Michels et al. [24]; Choe et al. [25] Autonomous robots: Abrams et al. [26] UTAUT was developed from TAM, TPB, and other models to AVs: Chan and Lee [21]; Adnan et al. [27]; Hewitt et al. [28]; Unifed theory of acceptance and use of technology explore user acceptance and includes four constructs (i.e., Meyer-Waaeden and Cloarec [29]; Zhang et al. [30] model (UTAUT) performance expectancy, efort expectancy, social infuence, UAVs: Kim and Chung [31]; Ganjipour and Edrisi [32] and facilitating conditions) [21] ADVs: Kapser and Abdelrahman [16]; Kapser et al. [33] CAT holds that emotions are aroused due to the individual’s Cognitive appraisal theory (CAT) evaluation of the stimulus, which further determines the AVs: Ribeiro et al. [34] behavioral response [34] MCI emphasizes the efect of motivation on customers’ AVs: Hewitt et al. [28] behaviors. Motivation is classifed into four main types, namely, Motivated consumer innovativeness (MCI) functional motivation, hedonical motivation, cognitive UAVs: Mathew et al. [1] motivation, and social motivation [1] TPB is designed to predict customers’ behavioral intentions Teory of planned behavior (TPB) according to three main components, namely, attitude, UAVs: Ramadan et al. [35] subjective norms, and perceived behavioral control [35] DOI is used to explain why customers adopt new technologies Difusion of innovation theory (DOI) and includes fve constructs (i.e., relative advantage, UAVs: Yoo et al. [14] complexity, compatibility, observability, and trialability) [14] 4 Journal of Advanced Transportation respondents showed a willingness to pay more, with one to accept the technology because respondent confdence levels often increase with such direct experience. group willing to pay up to $2.92 (average) or between 0-$6 for the service. Abrams et al. [26] proposed a new concept, “existence acceptance” to extend TAM. Existence accep- 2.1.2. Autonomous Delivery. Autonomous delivery includes tance refers to pedestrians’ or other trafc respondents’ not only ADVs but also UAVs and autonomous robots. acceptance of sharing the same roads with autonomous UAVs, also commonly referred to as drones, are fying robots. robots that can be remotely controlled or autonomously fown to deliver more light-weight, compact goods. Au- tonomous robots refer to robots that also operate and travel 2.1.3. Teoretical Background. In previous studies of au- upon sidewalks, usually complemented by a mothership van tonomous driving and autonomous delivery, TAM and that can transport the robots close to the delivery zone [7]. UTAUT are the most widely used behavioral theories for exploring new technology adoption (see Table 1). However, (1) UAVs. Combining TAM and DOI theory, Yoo et al. [14] TAM and UTAUT are not suitable for explaining customer adoption intention toward ADV services because customers proposed a model to investigate the factors that afect customer perceptions of UAVs and analyze the efect of are not the real users of ADVs. Te logistics practitioners are delivery area. Te researchers found that two major per- the direct users of ADV technology, not the customers, who ceived advantages of UAVs are their speed and environ- do not need to be familiar with the technology [17]. Te mental friendliness. In their study, perceived risk was also perceived usefulness and perceived ease of use are not the incorporated and defned to include performance risk, de- main concerns of customers when deciding whether to livery risk, and privacy risk. Based on customer motivated adopt ADVs in instant delivery. Tus, TAM and UTAUT innovativeness, Hwang et al. [44] applied consumer in- may be not well suited for examining and characterizing novativeness theory to study customer attitudes and ADV adoption intention. Another approach that could be considered to investigate technology adoption is Cognitive adoption intentions toward UAV food delivery. Te results showed that attitude played an important role in intention to Appraisal Teory (CAT). Tis theory holds that emotion is a kind of individual mental state that is generated when use and intention to pay more. Mathew et al. [1] explored customer intention to adopt UAVs for food delivery in evaluating relevant information and that infuences indi- India. Perceived risk and green image were found to be vidual behavioral motivation. One advantage of CAT is that signifcant predictors of customer attitude and intention. it can analyze customer behavior through a wide range of Aydin [45] investigated public knowledge and acceptance of emotional experiences. But at the same time, too many UAVs in specifc application areas, for example, commercial cognitive evaluation dimensions make it difcult for re- and public safety applications. However, the results showed searchers to capture quick and unconscious evaluation re- sults. Tis theory also does not answer the question that that UAVs were not generally accepted except for in the arenas of public safety and scientifc research. Need for whether any kind of emotion can be acquired without an evaluation system [48]. Motivated consumer innovativeness human interaction was added into the UTAUT2 model in Ganjipour and Edrisi’s study and was found to have (MCI) theory models motivation according to four main dimensions, namely, (1) functional, (2) hedonical, (3) cog- a negative efect on adoption intentions toward delivery drones [32]. Other studies used the Knowledge, Attitude, nitive, and (4) social [1]. In MCI, social motivation refers to and Practice Model or other methods to explore customer customers wanting to improve their image by purchasing acceptance [45–47]. innovative products/services (rather than referring to cus- tomers imitating the purchasing decisions of those in their (2) ADVs. Kapser and Abdelrahman [16] studied customer social networks). Teory of Planned Behavior (TPB), which intention toward ADVs and analyzed various key behavioral assumes that individuals consciously control their actions, has been used to study a wide range of health-related be- components (e.g., performance expectancy, efort expec- tancy, perceived risk, and price sensitivity) based on UTAUT haviors and learning behaviors. It includes three main variables, i.e., attitude, subjective norms, and perceived theory. Te results showed that respondents were neutral on the use of ADVs as a delivery alternative while price sen- behavior control. Although this theory has been applied by sitivity exerted a great infuence on customer intention. many researchers, there is a lack of agreement about its main Building on this research, Kapser et al. [33] integrated trust variables, raising doubts about the accuracy of results from and innovativeness into their acceptance model while factors research based on this theory. In addition, there are many shown to be insignifcant (i.e., efort expectancy) were re- behaviors that may not be explained by the main variables moved. In addition, a moderating factor, i.e., gender, was profered for this theory [49]. As an alternative, DOI theory explains how innovative technologies and ideas spread applied and they found that social infuence, perceived risk, and hedonic motivation had particularly signifcant efects through society over time and is also applied in commu- nication and information systems research to help explain on the use intentions of women customers. why customers adopt innovative technologies [14, 50]. Given its particular focus on innovative technology adoption, DOI (3) Autonomous robots. Pani et al. [11] analyzed customers’ willingness to pay more for autonomous robots by dividing theory is selected as the behavior theory of this study customer groups based on latent class analysis. 61.28% of of ADVs. Journal of Advanced Transportation 5 Table 2: Constructs used in DOI. Construct Main fndings References Duan et al. [62]; Ooi et al. [55]; Wang et al. [56]; Chung [57]; Kang et al. [58]; Relative advantage Relative advantage can promote adoption intention Hamad et al. [63]; Yoo et al. [14]; Min et al. [60] Complexity Complexity is a barrier to adoption intention Duan et al. [62]; Wang et al. [56]; Chung [57]; Yoo et al. [14]; Min et al. [60] Duan et al. [62]; Wang et al. [56]; Chung [57]; Kang et al. [58]; Hamad et al. [63] Compatibility Compatibility can promote adoption intention Min et al. [60]; Yoo et al. [14] Observability Observability can promote adoption intention Duan et al. [62]; Chung [57] Trialability Trialability can promote adoption intention Duan et al. [62]; Chung [57]; Hamad et al. [63] Social infuence Social infuence can promote adoption intention Min et al. [60] Green image/environmental Green image/environmental concern can promote adoption Yoo et al. [14] concern intention Perceived risk Perceived risk is a barrier to adoption intention Yoo et al. [14] 6 Journal of Advanced Transportation adoption intentions toward ADVs in China by adding the 2.2. DOI Teory Models of User Perception toward a New Technology. DOI theory as presented by Rogers [51] em- infuences of use experience, infection risk, and social awkwardness to DOI theory. ploys more specifc innovation characteristics, which are not typically found in behavioral theories in order to explain the adoption of new technologies or products. According to 3. Methodology Rogers, “the innovation is an idea, practice, or object that is In this section the constructs that are used in this study’s perceived as new by an individual or other unit of adoption.” customer adoption intention model are identifed and the Rogers also contends that innovations are difused through model structure is described. Customer adoption in- various channels and within a particular social system. tention here refers to the degree of willingness to use Rogers [52] presented fve innovation characteristics, ADVs (i.e., intention to use) and the willingness to pay namely, (1) relative advantage, (2) complexity, (3) com- more for the use of ADVs (i.e., intention to pay more). patibility, (4) observability, and (5) trialability. Relative First, the conceptual research model, which includes DOI advantage is defned as “the degree to which an innovation is theory supplemented with three new constructs, is pre- perceived as better than the idea it supersedes.” Complexity sented. Te proposed “Customer Adoption Intention is defned as “the degree to which an innovation is perceived toward Autonomous Delivery Vehicle (CAI-ADV)” as difcult to understand and use.” Compatibility involves model is depicted in Figure 1. Next, the survey design is “the degree to which a service is perceived as consistent with delineated, including the time, place, and scale of the users’ existing values, beliefs, habits, and present and pre- questionnaire, as well as the questionnaire’s content. vious experiences.” Observability is defned as “the degree to Finally, the data analysis process is presented. In addition which the innovation can be visible to others.” Finally, to the constructs, demographic factors as control vari- trialability is defned as “the extent to which the innovation ables are considered in our analyses. can be trialed, modifed, and experienced before the adoption” [53]. However, previous studies found that only three characteristics—relative advantage, complexity, and 3.1. Constructs in the Conceptual Research Model and Pro- compatibility—impact innovation adoption [14, 50, 54–62]. posed Hypotheses. Te constructs are divided into two parts, DOI theory has proven to be an efective theory for exploring namely, the original constructs from DOI and constructs customers’ acceptance of a new technology. Constructs used that we added. Te former includes relative advantage, in the original formulation of DOI theory and added in other complexity, and compatibility. Te latter encompass social subsequent studies are reviewed in Table 2. As indicated in infuence, perceived risk, green image, infection risk, social Table 2, relative advantage, complexity and compatibility are awkwardness, and use experience. Defnitions, related re- the most frequently applied constructs of DOI. Social in- search, and our hypotheses for each construct are introduced fuence has moreover been found to promote the adoption of as follows: new technologies [60]. Green image and perceived risk have Relative advantage (RA) is defned as “the degree to also exhibited positive efects on adoption intention which an innovation is perceived as better than the idea [1, 14, 18]. it supersedes” [53]. Previous research has found RA to be a positive predictor of behavioral/use intention [14, 55–58, 60, 62, 63]. In this study, relative advantage is 2.3. Research Gap. A review of the literature in the auton- defned as a customer’s subjective evaluation of the omous driving and autonomous delivery felds revealed only benefts aforded by ADVs against those by courier two studies [16, 33] that appear to explicitly concentrate on delivery. Autonomous delivery typically enables cus- the acceptance of ADVs. In contrast with these two studies, tomers to schedule delivery of their orders at any a time which were based on UTAUT theory, the DOI theory of their choice [64]. Customers can place orders at employed in this study focuses more on how innovative midnight and be served by ADVs without worrying technologies spread in society rather than on the use of those about order rejection. Time convenience is an im- technologies. We also incorporate some new constructs that portant characteristic of ADVs, making it a particularly may afect the ADV adoption intention of end customers, attractive delivery option for customers. Moreover, such as use experience and infection risk. As ADVs have without the personal preferences of couriers hampering become regularly available as a delivery option in Anting, options, orders from remote areas would not be Shanghai, some respondents to our survey have already used rejected and wait times may be greatly reduced during ADVs, with their use experiences increasing their confdence inclement weather. Time convenience [58] and zero in ADV services. Te risk of COVID-19 infection caused by rejection rates in harsh conditions may result in more person-to-person contact is likely to encourage customers to adoption. In addition, there are reports of couriers more readily adopt ADV supported services due to the contaminating customer food orders [65]. In contrast contactless delivery that ADVs enables. Tis study also to courier delivery, delivery by ADVs could better considers the impacts of social awkwardness on ADV protect the cleanliness and integrity of goods. Tere- adoption intention as ADVs can bypass direct communi- fore, the following hypothesis is proposed: cations with couriers and may become more popular among customers who are not so adept at socializing. To the best of H1: Relative advantage is positively related to customer our knowledge, this study is the frst to investigate customer intentions toward using ADV services. Journal of Advanced Transportation 7 Compatibility Relative advantage Complexity H2 H3 H1 Social influence Infection risk H7 H4 Customer H5 H8 adoption Perceived risk Social awkwardness intention H6 H9 Use experience Green image Income Education Age Gender Constructs from DOI Added Constructs Control variables Figure 1: Customer adoption intention toward autonomous delivery vehicle model (CAI-ADV). Complexity is defned as “the degree to which an in- home. Such an alteration engendered by the new de- livery mode demonstrates a gap in compatibility as novation is perceived as difcult to understand and use” [53]. Duan et al. [62], Wang et al. [56], Chung [57], compared to the customer’s usual delivery demand. As Yoo et al. [14], and Min et al. [60] found that com- a result, the customer’s adoption intention is lower. On plexity had a negative infuence on adoption of a new the other hand, if ADV service is not much diferent technology. In our study, complexity refers to the from courier delivery experiences, such as the ADVs problems that may arise in relation to a customer delivery price being the same as the previous delivery getting a notifcation when an order has arrived and/or price, customers’ ADV adoption intention would be difculties possibly experienced by a customer in re- higher. Terefore, the following hypothesis is proposed: trieving the food from the vehicle. For example, the H3: Compatibility is positively related to customer operation interface is challenging for some elderly intentions toward using ADV services. people to use, especially for picking up goods through Social infuence can be described as the phenomenon in a verifcation code. Without person-to-person contact which “people adopt when enough other people in the with a courier, it can be more difcult to return and group have adopted, that is, innovations spread by exchange goods. Terefore, the following hypothesis is a conformity motive” [66]. In other words, people will proposed: consciously imitate behaviors as exhibited in their H2: Complexity is negatively related to customer in- social network. Consistent with Higgins et al. [67], Min tention toward using ADV services. et al. [60] also found that social infuence had a positive efect on the acceptance of a new technology. Social Compatibility is defned as “the degree to which infuence is also considered in this study in order to a service is perceived as consistent with users” existing explore the positive feedback and recommendations values, beliefs, habits, and present and previous ex- communicated between members of close social groups periences” [53]. Studies from Duan et al. [62], Wang et as they relate to intentions toward ADV adoption. al. [56], Chung [57], Kang et al. [58], Hamad et al. [63], Terefore, the following hypothesis is proposed: and Min et al. [60] show that compatibility can posi- H4: Social infuence is positively related to customer tively afect customers’ intentions. Whether ADV intentions toward using ADV services. service is compatible with customers’ usual delivery demands will afect their intentions toward using Perceived risk has been generally found to be a useful ADVs. In this study, compatibility refers to the con- addition to DOI in studying customer intentions to- sistency between the new ADV service and some ward a new technology. Yoo et al. [14] and Mathew et previous experience with or more habitual use of in- al. [1] both used three dimensions to characterize perceived risk, performance risk, delivery risk, and stant delivery by customers. For example, ADVs cannot provide home delivery services in some residential privacy risk. ADV technology has the potential to al- leviate the problem of courier accidents. Whether areas or ofce buildings. A customer that used to enjoy door-to-door delivery service may then need to pick up ADVs can complete the delivery process and avoid goods delivered by ADVs in a certain area such as trafc accidents (i.e., delivery risk) is a thought-pro- a parcel locker that is three hundred meters away from voking problem. ADVs have not been implemented 8 Journal of Advanced Transportation the less intention there is to try drone delivery [32]. extensively enough to accurately evaluate their ability to reduce accidents. Due to the need to draw on a large Drawing upon this fnding, customers who are socially awkward, meaning that they are less adept at or in- amount of sensitive data online such as names, ad- dresses, and phone numbers, protecting privacy is clined to communicate with other people, may fnd another important concern that may afect customer ADVs to be an attractive option. Terefore, the fol- intentions. Furthermore, network instability can also lowing hypothesis is proposed: afect the operation of ADVs. An additional technical H8: Social awkwardness is positively related to cus- risk is hacking (e.g., breach of frewalls). Considering tomer intentions toward using ADV services. these risks, the more they are reduced, the higher the Studies from Bernhard et al. [41] show that use ex- adoption intentions of customers. Previous studies perience can afect customers’ intentions because found that perceived risk had a negative efect on ac- customers respond on the basis of real experience, not ceptance of a new technology [1, 14]. Terefore, the imagination. However, few studies have taken this perceived risk discussed in this study refers to the construct into consideration. Customers who have used delivery risk, privacy risk, and technical risk. In light of ADVs are better able to understand ADV delivery these considerations, the following hypothesis is service, including the process, its advantages, and proposed: disadvantages. Customer confdence in the service H5: Perceived risk is negatively related to customer could then increase as a result of direct experience. intentions toward using ADV services. Based on previous experiences, customers may be more likely to choose ADVs as a delivery option. Terefore, Because of global warming and increasing air pollution, the following hypothesis is proposed: the environmental impacts of any given transportation system has become a factor worth considering [68]. Te H9: Use experience is positively related to customer emergence of autonomous vehicles powered by elec- intentions toward using ADV services tricity enables cleaner energy consumption [64]. Te implementation of ADVs is expected to be a valuable step towards zero-emission logistics in the United 3.2. Survey Design. Based on the research model proposed in States [69]. Many studies show that customers’ envi- Section 3.1, we designed and conducted an online and ofine ronmental concerns could afect their adoption in- survey to test the proposed hypotheses. Our online survey tentions toward green products such as Avs and UAVs was conducted through the Sojump website, and the ofine [1, 14, 18]. In this study, green image is defned as survey was conducted in Anting, Shanghai. In Anting, the “customers” perceptions of ADVs having a salutary Yonghui supermarket began implementing ADV service in environmental performance.” Terefore, the following 2020, covering a delivery area of 5 square kilometers with hypothesis is proposed: more than six thousand households [71]. Te size of a sample from a given population necessary to H6: Green image is positively related to customer in- achieve a good or very good degree of adequate represen- tentions toward using ADV services. tativeness has been suggested to be 300 or 500, respectively Most countries around the world have been greatly [72] or alternatively achieved through a minimum 20 :1 ratio afected by COVID-19. Te disease spreads primarily between the sample size and the number of model pa- through contact with an infected person (even through rameters to be statistically estimated [73]. Tus, the 691 valid a cough or sneeze) or touching a surface that has the samples collected for our survey data analysis satisfy both of virus on it. Due to the fear of infection, contactless these recommendations. Te data collection took place in delivery (as enabled through ADVs and UAVs) has October and November in 2021. become increasingly popular, especially in areas ex- An introduction to ADVs is presented at the beginning periencing high infection risk. Te chance of human- of the survey, including history, delivery process, equipment to-human transmission during the COVID-19 pan- parameters, advantages, and disadvantages, in order to demic is reduced through the implementation of provide a clearer understanding of ADVs for respondents. ADVs. Terefore, the following hypothesis is proposed: Te second part of the survey includes questions about H7: Infection risk is positively related to customer respondents’ intention to adopt ADV services. Tese intentions toward using ADV services. questions are developed with reference to previous relevant Te results of Xi et al.’s study [70] show that same-day- literature. Consistent with previous surveys [1, 16, 26, 44], delivery has a negative efect on greeting/chatting with constructs except for use experience are measured based on neighbors but has a positive association with social a seven-point Likert-type scale, ranging from strongly dis- gatherings that often occur between acquaintances. agree (1) to strongly agree (7). Based on the hypotheses ADV technology can make our lives more convenient proposed in 3.1, the related observational variables and but can also further alienate us from our neighbors or corresponding references are presented in Table 3. Te third other casual acquaintances. Similar to autonomous part of the survey includes basic individual information (e.g., taxis [22], ADVs circumvent the person-to-person age, gender, monthly household disposable income, edu- contact that comes with using couriers. One study has cation, job, and instant delivery experience). To improve the found that the greater the need for human interaction, validity of the questionnaire, reverse worded questions are Journal of Advanced Transportation 9 Table 3: Constructs, observational variables, and supporting references. Construct Item Source RA1: Using autonomous delivery vehicles would help me receive orders more quickly Relative advantage (RA) RA2: Autonomous delivery vehicles can operate in inclement weather Mathew et al. [1] RA3: Autonomous delivery vehicles can deliver at any time I want RA4: Using autonomous delivery vehicles would be cool COL1: Interacting with autonomous delivery vehicles is clear and understandable COL2: I believe that it is easy to make autonomous delivery vehicles to do what I Complexity (COL) Yoo et al. [14]; Wu et al. [18] want COL3: I believe that retrieving goods from autonomous delivery vehicles is easy COA1: Using autonomous delivery vehicles is compatible with my past experience and habitual use of delivery service Compatibility (COA) Yoo et al. [14] COA2: Using autonomous delivery vehicles is comparable to the pickup distance and delivery price of my previous experience SI1: I can be infuenced by my peers’ preferences or decisions in choosing whether to use autonomous delivery vehicles Social infuence (SI) Hamad et al. [63] SI2: Te more people who have used autonomous delivery vehicles, the more willingness to use I have PR1: Te package the vehicle is carrying might be damaged or stolen PR2: Te delivery process may take too long PR3: Autonomous delivery vehicles will cause disclosure of private information such as address, name, and telephone number Perceived risk (PR) Yoo et al. [14]; Mathew et al. [1] PR4: Te remote-control center and the operating system of autonomous delivery vehicles may break down PR5: Hackers may attack the system and steal personal data, even operate autonomous delivery vehicles to attack pedestrians GI1: Using autonomous delivery vehicle services can reduce environmental pollution Green image (GI) Mathew et al. [1]; Li et al. [22] GI2: By using autonomous delivery vehicle services, I can demonstrate that I care about environmental conservation Infection risk (IR) IR: Using autonomous delivery vehicles can reduce the infection risk Tis study SA1: When communicating with strangers, I tend to feel uncomfortable Social awkwardness (SA) SA2: Using an autonomous delivery vehicle can remove the need for Tis study communication with couriers Use experience (UE) UE: Have you used autonomous delivery vehicles before? Tis study AI1: I Intend to use autonomous delivery vehicles as a delivery option in the future Adoption intention (AI) Kapser and Abdelrahman [16]; Hwang et al. [44] AI2: I am willing to pay more for autonomous delivery vehicle services 10 Journal of Advanced Transportation included in this survey. Responses containing what are 44.86% of them. Respondents with only a high school degree functionally contradictory “strongly agree” answers to such also account for the largest portion of respondents. Te questions, along with responses that were completed in less demographic analysis reveals that the sample is also rep- than 50 seconds, are excluded from analysis. resentative for the Chinese population in terms of age and gender [78]. Terefore, it is possible to generalize the fndings to a larger population in China. Te number of 3.3. Data Analysis. Our analysis includes two approaches, respondents who have used instant delivery service and namely, demographic analysis and structural equation those who have not (50.65% and 49.35% respectively) are modeling (SEM). Te demographic analysis includes a basic almost half and half. Takeaway food delivery is used most, profle of customers and the degree of infection risk for the followed by fresh food delivery and medicine delivery. It is location under study. Tese demographic characteristics worth noting that 56 respondents (8.1% of the survey such as age, gender, income, and education are analyzed as sample) indicated that they already have experience using control variables to capture their efects on customers’ ADV services. At the same time, as shown in Figure 2, adoption intentions. Te demographic analysis is followed 58.32% respondents are not willing to pay extra for ADV by SEM, which can be used to study the relationships be- service. Finally, the results shown in Figure 3 indicate that tween observed variables and latent variables [74]. SEM 58.93% of respondents remain neutral toward ADVs, which includes two stages-measurement model assessment and is consistent with previous research [16]. structural model assessment. Te measurement model assessment is performed to test 4.2. Measurement Model Assessment. Both the Cronbach’s the reliability and validity of the questionnaire. To validate alpha and CR for all the items of this study are found to be the model construct, four steps are implemented. Te frst above 0.70, except the Cronbach’s alpha for AI (Table 5). step is to test the internal construct reliability with Cron- According to a study by McCrae et al. [79], the reason for bach’s alpha and the composite reliability (CR), the outputs this exception may be the low number of questions re- of which should have thresholds higher than 0.70 [75, 76]. garding AI in the survey [80]. Te AVE values range from Second, convergent and discriminant validity should both be 0.612 to 0.926, which are all higher than 0.5, thus establishing examined. Convergent validity is assessed in terms of av- validity. Factor loadings range from 0.669 to 0.972 are all erage variance extracted (AVE) with a threshold of 0.50. In also higher than 0.5 [76] (Table 5). Te results show that addition, factor loadings should be higher than 0.5 [76]. convergent validity is established, indicating good agree- Tird, discriminant validity should be examined to measure ment between two or more items that measure the same how a construct is diferent from other constructs in the structure. Although the Cronbach’s alpha for AI is below SEM model by calculating the degree to which it correlates 0.70, the rest of parameters of AI including factor loadings, with other constructs and how distinctly it exists as a unique CR, and AVE all meet the standard, thus subsequent analysis construct [76]. In order to verify good discriminative val- can be conducted. √�� idity, the square root of AVE should be greater than the Te value of AVE is presented on the diagonal in correlation coefcient between the construct and other Table 6 and correlation coefcients are presented below the √�� constructs [76]. Finally, confrmatory factor analysis is used diagonal. Each variable has a greater AVE than the cor- to assess the measurement model [73]. Te parameters in relation coefcient, which demonstrates discriminant val- confrmatory factor analysis and their standards are drawn idity, i.e., that each construct is distinct from each other. from Cheung et al. [77]. Te results show that the proposed model has a greater Te structural model assessment is employed for path ft to the data based on the standards showed in Table 7. analysis. Te results include path coefcients and p-values. Only when the p-value <0.05 is the construct deemed to have a signifcant infuence on intention, thus supporting the 4.3. Structural Model Assessment. Te measurement model corresponding hypothesis. AMOS version 24 (Asymptoti- assessment is followed by structural model assessment cally distribution-free) is used for data analysis. [73, 76]. Te results, including path coefcients and p -values, are shown in Table 8 and Figure 4. Only when the p -value <0.05 is when the construct deemed to have a sig- 4. Results and Discussion nifcant infuence on intention, thus supporting the corre- sponding hypothesis. Te analysis from Table 8 shows that 4.1. Demographic Analysis. Te demographic details of re- spondents are shown in Table 4. Te percentages of re- eight of nine hypotheses are supported. More specifcally, compatibility (β � 0.171, p< 0.001), social infuence (β spondents who are male and female are 42.40% and 54.85%, respectively. Te percentage of respondents aged between � 0.205, p< 0.001), infection risk (β � 0.076, p< 0.05), green 25–34 and 35–49 years are 23.44% and 36.61%, respectively. image (β � 0.229, p< 0.001), social awkwardness (β � 0.302, Te distribution of other age groups generally mirrors that of p< 0.001), and experience (β � 0.040, p< 0.05) have a posi- the population studied, whereas the percentage of elderly tive infuence. Complexity (β � −0.294, p< 0.001) and per- over 65 is rather low due to the lower participation of the ceived risk (β � −0.149, p< 0.001) have a negative infuence elderly in using intelligent devices. Respondents with av- on customers’ intentions. Hence, hypotheses 2, 3, 4, 5, 6, 7, 8, and 9 are supported. However, contrary to expectation, erage monthly disposable incomes of 2000–6000 RMB make up the largest proportion of respondents, accounting for relative advantage (β � −0.058, p � 0.195) does not have Journal of Advanced Transportation 11 Table 4: Demographic details of respondents. Variable Category Frequency Percentage (%) Female 379 54.85 Gender Male 293 42.40 Tird gender 19 2.75 Below 18 years 115 16.64 19–24 96 13.89 25–34 162 23.44 Age 35–49 235 36.61 50–64 75 10.85 Above 65 years 8 1.16 Below 2000 RMB 143 20.69 2000–6000 RMB 310 44.86 6000–10000 RMB 120 17.37 Monthly household disposable income 10000–15000 RMB 58 8.39 15000–20000 RMB 21 3.04 Above 20000 RMB 39 5.64 Secondary school certifcate or below 28 4.05 High school degree 241 34.88 Bachelor’s degree 133 19.25 Education Master’s degree 213 30.82 Doctorate 58 8.39 No degree 18 2.60 Yes 350 50.65 Instant delivery use experience No 341 49.35 Fresh food 155 22.43 Take-away food 285 41.24 Instant delivery services used Medicine 89 12.88 Other 42 6.08 Yes 56 8.10 ADV use experience No 635 91.90 Medium and high-risk area 23 3.33 Epidemic risk level Low-risk area 668 96.67 a signifcant positive infuence. Te practical reason may be processes supporting the pickup and return of goods, which that the relative advantage in this context relates to whether are major components of complexity in the context of ADV ADVs can complete the delivery process and meet the de- service, should be as clear and efcient as possible to maximize customer ease of use of the service. An example livery demands of customers, while how goods are delivered is not the main concern of customers. Tis view was also process would be “face swiping,” which has become widely applied in check-in and -out services in Chinese railway expressed by many respondents during the ofine investigation. stations for picking up and returning goods. In addition to the variables related to our hypotheses, the In line with fndings from Yoo et al. [14], Min et al. [60], infuence of control variables (i.e., age, gender, income, and and Hamad et al. [63], our results show that compatibility education) on adoption intention is also studied (Table 9). has a positive efect on customers adoption intentions to- Except for gender, the control variables that were included in ward ADV services. Tis indicates that when ADV service is our survey are found to have signifcant efects on customers’ more consistent with ’customers’ previous delivery experi- intentions, although the efects sometimes diverge in direction. ence and habits, ’customers’ adoption intentions are higher. In addition, in comparison with the previous results without For example, traditional courier service enables goods to be delivered directly to one’s doorstep, but ADVs may not be control variables, the inclusion of age, gender, and income does not change the signifcance of any of the constructs (see Table 8). allowed to enter a given neighborhood or are not able to Tus, it is demonstrated that the model has good robustness. climb stairs. In these conditions, customers need to go Te results of path analysis provide evidence supporting downstairs to pick up goods. Terefore, some management DOI in the context of ADVs. Te results also indicate that regulations may need to be established to allow ADVs to infection risk, use experience, and social awkwardness are operate in particular residential areas, and ADVs may need important constructs afecting customer adoption intentions. to be equipped with RFID technology to communicate with the lift or parcel locker. Tus, home delivery can be ofered 4.4. Discussion. In accordance with fndings from previous and the convenience of using ADVs is improved. Kapser and studies [14, 60], complexity has a negative efect on cus- Abdelrahman [16] found that when the price is higher than tomers’ intention to adopt ADVs in instant delivery. Te the current price of courier delivery, customers are much less 12 Journal of Advanced Transportation Table 5: Results for factor loading, Cronbach’s alpha, CR, AVE, Extra cost respondents could accept and Sqrt of AVE. 1.74% 6.80% √�� Factor Cronbach’s Variable Item CR AVE AVE loading alpha RA1 0.905 RA2 0.905 RA 0.899 0.938 0.790 0.889 RA3 0.897 38.49% 52.97% RA4 0.848 COL1 0.919 COL COL2 0.957 0.928 0.959 0.885 0.941 COL3 0.946 COA1 0.958 COA 0.933 0.961 0.926 0.962 COA2 0.966 SI1 0.880 SI 0.724 0.905 0.827 0.909 SI2 0.938 No 10-20% PR1 0.916 <10% >20% PR2 0.918 Figure 2: Percentage of respondents according to acceptable in- PR PR3 0.900 0.905 0.969 0.863 0.929 crease in cost for ADV service. PR4 0.959 PR5 0.950 GI1 0.949 Attitude towards ADVs GI 0.934 0.960 0.923 0.961 GI2 0.972 4.49% SA1 0.951 SA 0.870 0.960 0.923 0.961 SA2 0.970 37.19% AI1 0.881 AI 0.573 0.759 0.612 0.782 AI2 0.669 Notes. RA � relative advantage; COL � complexity; COA � compatibility; 58.32% SI � social infuence; PR � perceived risk; GI � green image; SA � social awkwardness; AI � adoption intention. Table 6: Results for discriminant validity (Fornell-Larcker criterion). Positive RA COL COA SI PR GI SA Neutral RA 0.889 Negative COL −0.830 0.941 COA 0.756 −0.816 0.962 Figure 3: Percentage of respondents according to attitude towards SI 0.631 −0.698 0.672 0.909 ADVs. PR 0.321 −0.334 0.316 0.474 0.929 GI 0.584 −0.651 0.656 0.675 0.348 0.961 likely to accept ADVs. To increase customer use intentions, SA 0.340 −0.363 0.388 0.394 0.536 0.424 0.961 it is recommended that the price be equal to or lower than Notes. RA � relative advantage; COL � complexity; COA � compatibility; courier delivery. Tese fndings are also consistent with our SI � social infuence; PR � perceived risk; GI � green image; SA � social awkwardness. results as shown in Figure 2, with more than 90% of re- spondents indicating that they are not willing to pay more than 10% for ADV service compared to courier delivery. Table 7: Results of confrmatory factor analysis. Moreover, the ADVs should be available for service late at night and during severe weather like couriers. Te ADVs CMIN/ Indices χ df RMSEA NFI GFI AGFI should have the ability to operate in such settings, such as DF being equipped with specifc sensors and waterproof casing. Standards — — <3 <0.08 >0.7 >0.9 >0.7 Te results also confrm a positive efect of social network in Results 567.486 253 2.243 0.037 0.825 0.934 0.893 the context of ADVs, congruent with the fndings of Higgins et al. [67] and Min et al. [60]. Once one person starts using ADVs, more and more of their friends and family also begin to as issuing discount coupons for ADV services that can be shared with their friends to encourage them to try ADV services. We use ADVs, with such use difusing along their social networks. Terefore, the infuence of social networks can be utilized to suggest that ADV service providers invite individuals who have strong social infuence to try ADV services, similar to the efect promote the expansion of ADV service. For example, ADV service providers can introduce some incentive strategies such of celebrity product endorsements in advertisements. Journal of Advanced Transportation 13 Table 8: Results of path analysis. Hypothesis Path Proposed efect β Signifcance Result H1 RA⟶ AI + −0.058 0.195 Rejected H2 COL⟶ AI − −0.294 <0.001 Supported H3 COA⟶ AI + 0.171 <0.001 Supported H4 SI⟶ AI + 0.205 <0.001 Supported H5 PR⟶ AI − −0.149 <0.001 Supported H6 GI⟶ AI + 0.229 <0.001 Supported H7 IR⟶ AI + 0.076 <0.05 Supported H8 SA⟶ AI + 0.302 <0.001 Supported H9 UE⟶ AI + 0.040 <0.05 Supported Notes. RA � relative advantage; COL � complexity; COA � compatibility; SI � social infuence; PR � perceived risk; GI � green image; IR � infection risk; SA � social awkwardness; UE � use experience; AI � adoption intention. Relative advantage Complexity Compatibility 0.171* -0.294* -0.058 Social influence Infection risk 0.076* 0.205* Customer 0.302* -0.149* adoption Perceived risk Social awkwardness intention 0.040* 0.229* Green image Use experience 0.054* -0.014 0.058* -0.058* Age Gender Income Education Constructs from DOI Added Constructs Control variables Figure 4: Results of path analysis. Notes: means that this attribute is a signifcant contributor according to the analysis. Table 9: Control variable results. surrounding environment, so that the customer can track the ADVs in real-time via an application. Second, service Path β Signifcance providers should strengthen the robustness of their re- Age⟶ AI −0.058 <0.001 spective systems. For example, after an attack perpetrated by Gender⟶ AI −0.014 0.475 hackers, a backup system prepared in advance could be used Income⟶ AI 0.058 <0.01 Education⟶ AI 0.054 <0.05 to ensure uninterrupted service. Tird, with reference to the (H.R.3711) SELF DRIVE Act of America [81], ADV man- ufactures should establish a clear and reasonable data Along with the other positive fndings of this study, protection plan that encompasses the collection, use, a negative efect from perceived risk was also found, in line sharing, and storage of information about vehicle owners or with previous studies [1, 14]. In our study, perceived risk, customers, in order to enhance customer confdence in data including delivery, privacy, and technical risk, are found to protection. exert a negative efect on customer adoption intentions. Due Infection risk was rarely considered in previous studies. In to the need to upload a large amount of data such as ad- the present study, it has proven to be an important positive dresses online to successfully provide ADV services, cus- predictor of customer intentions. Contactless delivery has tomers are sensitive about protecting their privacy. Because become an efective method for reducing the spread of COVID-19 and other diseases while meeting daily needs. of the usage of the Internet in the process of delivery, system safety and robustness are also important components that ADVs play an important role in epidemic prevention and control. We suggest that more ADVs be deployed in medium- should be considered to ensure the smooth operation of the whole process and prevent hacking. To address these con- high infection risk areas, including hospital settings. cerns, we recommend the following three measures: First, Green image also demonstrated a signifcant positive the ADV should be installed with high-precision mapping efect on customer intentions, in agreement with previous and detection equipment to improve awareness of the studies [1, 14, 18]. Tese fndings indicate that protecting the 14 Journal of Advanced Transportation could be ofered to promote its use among those who can environment has become increasingly important to cus- tomers. ADVs may be an especially attractive option given serve as early adopters. that they are powered by electricity and have a lower adverse impact on the urban environment as compared to the use of 5. Conclusion delivery trucks and motorcycles. We suggest that the adoption of ADVs could be accelerated by promoting their In this study, we propose a customer adoption model that superior environmental performance as compared to the applies a DOI theory model with three new constructs and traditional option. Nonetheless, it should be acknowledged that focuses on customer adoption intention toward ADV that vehicle electrifcation and the carbon intensity of services in instant delivery. To our knowledge, this is the frst electricity still have key infuences on environmental study that considers impacts of use experience, infection performances [82]. risk, and social awkwardness in the context of ADVs. Eight Te results of the present study also indicate that social of nine constructs (i.e., complexity, compatibility, social awkwardness is an important positive predictor of cus- infuence, perceived risk, infection risk, green image, social tomers’ intentions. Perhaps due to the development of awkwardness, and use experience) are found to be statis- online shopping, social activities are decreasing, and people tically signifcant contributors to customer intentions to- are more accustomed to being alone or just communicating ward adopting ADV services. Furthermore, 58.32% of the with acquaintances. Direct contact with couriers may make respondents surveyed in this study expressed a neutral at- some customers feel awkward or even be perceived as po- titude toward ADV adoption, while 47.03% of respondents tentially dangerous, especially for women who live alone, expressed a willingness to pay more for ADV services thus making interactions with ADVs an acceptable or compared to courier delivery. Tis study also examines the a desirable alternative. Although social awkwardness pro- efects of age, gender, income, and education on adoption motes the use of ADVs, it may also aggravate the degree of intentions. Income and education are found to have sig- social isolation. To address this concern, ADV developers nifcant positive efects on the intention to adopt ADVs, can develop more humanized interactions between ADVs while age is found to serve as a negative predictor. Tese and humans during delivery services. results shed light on which socio-demographic factors exert Use experience is also found to positively afect customer the most infuence on adoption intentions towards ADV adoption intentions. Te results of path analysis confrm that service. customers who have already used ADVs are more likely to Overall, this study ofers some important insights into accept ADV services. Te reason may be that based on their the infuences of various constructs on customers’ intentions use experiences, their understanding of ADV services is to adopt ADVs in instant delivery. Tis study not only deeper and trust and confdence in ADVs services are in- enriches the research by providing a more refned theoretical creased by the direct use experience. Based on their previous framework for characterizing customer intentions toward experience with ADVs, they are more inclined to choose adopting ADV services, but also contributes to promoting ADVs as a delivery option. If more customers are provided the expansion of ADVs in instant delivery. It is the frst study use opportunities, for example through trial activities held in to investigate customer adoption intention toward ADVs some delimited areas such as industrial parks, then large- using DOI theory and broadening its application, rather scale use of ADV services could be achieved. than utilizing TAM theory, in recognition of the lack of end Lastly, age, income, and education are all found to be consumer concern for matters related to the direct use of signifcant contributors to customers’ adoption intentions. ADVs, with personal innovativeness and social infuence While age exhibited a negative efect on adoption intentions, being more prominent factors afecting customer intentions income and education both displayed positive efects. toward ADVs. New constructs are incorporated into the Compared to younger customers, older adults are more proposed model to adapt to the specifc context of ADVs, likely to be conservative and have less propensity to use new including infection risk, social awkwardness, and use ex- technologies. Tus, they are less willing to use driverless perience, with each found to be important predictors of vehicles for delivery. For customers with lower incomes, customer intention. In terms of practical implications, the a key reason that may discourage them from choosing ADVs fndings of this study can contribute to the formulation of is a concern about the delivery price. Tey may think that the relevant regulations (data privacy, security risks, etc.), delivery price of a new technology may be higher than the marketing strategies (such as discount coupons) and tech- traditional courier delivery. Tus, they are more likely to nology development (such as RFID and waterproofng) for choose courier delivery. Finally, customers with higher ADVs, furthering the popularization of their use. education may know more about ADVs. Terefore, they Te study is not without limitations. Te samples in- trust this new technology more and are more willing to volved in this study are all from China, so future research adopt ADVs. Based on the previous analysis, two sugges- should examine sample data covering diferent countries in tions are proposed as follows: (1) the use of ADVs could be order to validate the model proposed in this study and to promoted among young and/or more highly educated diferentiate geographical efects. In addition, only 8.1% of customers, so they can act as early adopters in their social respondents in our survey sample indicated prior ADV networks; with the assistance of such customers, older adults service use experience. Future research could investigate could gain more familiarity and comfort with ADV service customers with more experience with ADV service and processes; (2) a discount on the delivery price of the ADVs conduct multigroup analysis to compare adoption Journal of Advanced Transportation 15 [11] A. Pani, S. Mishra, M. Golias, and M. Figliozzi, “Evaluating intentions between non-users and users, identifying possible public acceptance of autonomous delivery robots during diferences in key variables such as the willingness to pay COVID-19 pandemic,” Transportation Research Part D: more. Another limitation is that it is a cross-sectional study Transport and Environment, vol. 89, 2020. that was conducted during the COVID-19 pandemic. [12] Udelv, “Autonomous delivery vehicles- why they matter, and Terefore, future investigation could include a longer in- how they work,” 2018. vestigation period and compare the results to postpandemic [13] Pengpai, “Meituan autonomous delivery vehicle crashed into periods. a private car, which was found to be entirely to blame: the ‘motor vehicle’ illegally entered a non-motorized lane,” 2021, Data Availability https://www.thepaper.cn/newsDetail_forward_14881840. [14] W. Yoo, E. Yu, and J. Jung, “Drone delivery: factors afecting Te sample data used to support the fndings of this study are the public’s attitude and intention to adopt,” Telematics and available from the corresponding author upon request. Informatics, vol. 35, no. 6, pp. 1687–1700, 2018. [15] Autocrypt, “When last Mile delivery turns autonomous – what are the considerations?,” 2021, https://autocrypt.io/ Conflicts of Interest when-last-mile-delivery-turns-autonomous/. [16] S. Kapser and M. Abdelrahman, “Acceptance of autonomous Te authors declare that there are no conficts of interest delivery vehicles for last-mile delivery in Germany – regarding the publication of this article. extending UTAUT2 with risk perceptions,” Transportation Research Part C: Emerging Technologies, vol. 111, pp. 210–225, Acknowledgments [17] B. V. Meldert and L. D. Boeck, “Introducing autonomous Tis work was supported in part by the National Natural vehicles in logistics a review from a broad perspective,” Report Science Foundation of China (72101188), the Shanghai KBI_1618, KU Leuven - Faculty of Economics and Business, Pujiang Program (2020PJC112), the Shanghai Municipal Leuven, Belgium, 2016. Science and Technology Major Project (2021SHZDZX0100), [18] J. Wu, H. Liao, J. W. Wang, and T. Chen, “Te role of en- and the Fundamental Research Funds for the Central vironmental concern in the public acceptance of autonomous Universities. Te authors would like to thank the editor and electric vehicles: a survey from China,” Transportation Re- reviewers for their insightful comments in improving the search Part F: Trafc Psychology and Behaviour, vol. 60, pp. 37–46, 2019. quality of this paper. [19] M. Liu, J. Wu, C. Zhu, and K. Hu, “Factors infuencing the acceptance of robo-taxi services in China: an extended References technology acceptance model analysis,” Journal of Advanced Transportation, vol. 2022, Article ID 8461212, 11 pages, 2022. [1] A. O. Mathew, A. N. Jha, A. K. Lingappa, and P. Sinha, [20] T. Dirsehan and C. Can, “Examination of trust and sus- “Attitude towards drone food delivery services—role of in- tainability concerns in autonomous vehicle adoption,” novativeness, perceived risk, and green image,” Journal of Technology in Society, vol. 63, Article ID 101361, 2020. Open Innovation: Technology, Market, and Complexity, vol. 7, [21] W. M. Chan and J. W. C. Lee, “5G connected autonomous no. 2, 2021. vehicle acceptance: the mediating efect of trust in the tech- [2] Foresight Research Institute, Analysis on the Market Status nology acceptance model,” Asian Journal of Business Research, Quo and Development Trend of China Instant Delivery In- vol. 11, no. 1, 2021. dustry in 2020, 2020. [22] H. Li, S. Yu, and J. Zheng, “Acceptance factors for younger [3] M. Lu, R. Wang, and P. Li, “Comparative analysis of online passengers in shared autonomous vehicles,” in Proceedings of fresh food shopping behavior during normal and COVID-19 the International Conference on Human-Computer In- crisis periods,” British Food Journal, vol. 124, no. 3, teraction, Virtual Event, July 2021. pp. 968–986, 2021. [23] S. Leon, C. Chen, and A. Ratclife, “Consumers’ perceptions of [4] P. Wang, X. Ma, and L. Hu, “Food delivery people become last mile drone delivery,” International Journal of Logistics a high incidence of trafc accidents,” 2020, http://www. Research and Applications, pp. 1–20, 2021. xinhuanet.com/mrdx/2020-09/15/c_139368949.htm. [24] M. Michels, C. F. von Hobe, P. J. Weller von Ahlefeld, and [5] AliResearch and Cainiao, National Socialized E-Commerce O. Musshof, “Te adoption of drones in German agriculture: Logistics Personnel Research Report, Jiaotong University, a structural equation model,” Precision Agriculture, vol. 22, Beijing, China, 2016. no. 6, pp. 1728–1748, 2021. [6] S. Chhabra, “Cost of last mile delivery for your business with [25] J. Y. Choe, J. J. Kim, and J. Hwang, “Innovative marketing ways to optimise it,” 2021, https://jungleworks.com/cost-of- strategies for the successful construction of drone food de- last-mile-delivery/. livery services: merging TAM with TPB,” Journal of Travel & [7] M. A. Figliozzi and D. Jennings, “A study of the competi- tiveness of autonomous delivery vehicles in urban areas,” Civil Tourism Marketing, vol. 38, no. 1, pp. 16–30, 2021. [26] A. M. H. Abrams, P. S. C. Dautzenberg, and C. Jakobowsky, and Environmental Engineering Faculty Publications and Presentations, 2020. “A theoretical and empirical refection on technology ac- ceptance models for autonomous delivery robots,” in Pro- [8] 36Kr, “Research report on autonomous delivery feld,” 2018, https://www.36kr.com/p/1725163094017. ceedings of the 2021 ACM/IEEE International Conference on Human-Robot Interaction, pp. 272–280, Boulder, CO, USA, [9] Flsenate, “CS/CS/HB 1289: autonomous vehicles,” 2021, https://fsenate.gov/Session/Bill/2021/1289. March 2021. [27] N. Adnan, S. Md Nordin, M. A. bin Bahruddin, and M. Ali, [10] Flsenate, “CS/CS/SB 1620: autonomous vehicles,” 2021, https://www.fsenate.gov/Session/Bill/2021/1620. “How trust can drive forward the user acceptance to the 16 Journal of Advanced Transportation technology? In-vehicle technology for autonomous vehicle,” [43] J. Berrada, I. Mouhoubi, and Z. Christoforou, “Factors of Transportation Research Part A: Policy and Practice, vol. 118, successful implementation and difusion of services based on pp. 819–836, 2018. autonomous vehicles: users’ acceptance and operators’ [28] C. Hewitt, I. Politis, and T. Amanatidis, “Assessing public proftability,” Research in Transportation Economics, vol. 83, perception of self-driving cars,” in Proceedings of the 24th Article ID 100902, 2020. International Conference on Intelligent User Interfaces, [44] J. Hwang, J. J. Kim, and K. W. Lee, “Investigating consumer pp. 518–527, Marina del Ray, CA, USA, March 2019. innovativeness in the context of drone food delivery services: [29] L. Meyer-Waaeden and J. Cloarec, “‘Baby, you can drive my its impact on attitude and behavioral intentions,” Techno- car’: psychological antecedents that drive consumers’ adop- logical Forecasting and Social Change, vol. 163, Article ID tion of AI-powered autonomous vehicles,” Technovation, 120433, 2021. vol. 109, 2021. [45] B. Aydin, “Public acceptance of drones: knowledge, attitudes, [30] S. Zhang, P. Jing, and G. Xu, “Te acceptance of independent and practice,” Technology in Society, vol. 59, 2019. autonomous vehicles and cooperative vehicle-highway au- [46] L. K. Lin Tan, B. C. Lim, and G. Park, “Public acceptance of tonomous vehicles,” Information, vol. 12, no. 9, 2021. drone applications in a highly urbanized environment,” [31] K. B. Kim and B. G. Chung, “Technology acceptance of in- Technology in Society, vol. 64, 2021. dustry 4.0 applying UTAUT2: focusing on AR ang drone [47] X. Zhu, T. J. Pasch, and A. Bergstrom, “Understanding the services,” Journal of Information Technology, vol. 26, no. 6, structure of risk belief systems concerning drone delivery: pp. 29–46, 2019. a network analysis,” Technology in Society, vol. 62, Article ID [32] H. Ganjipour and A. Edrisi, “Applying the integrated model to 101262, 2020. understanding online buyers’ intention to adopt delivery [48] M. Yarwood, Psychology of Human Emotion. Afordable drones in Iran,” Transportation Letters, vol. 15, no. 2, Course Transformation, Pennsylvania State University, pp. 98–110, 2022. Pennsylvania, PA, USA, 2022. [33] S. Kapser, M. Abdelrahman, and T. Bernecker, “Autonomous [49] W. Duan and G. Jiang, “A review of the theory of planned delivery vehicles to fght the spread of Covid-19 - how do men behavior,” Advances in Psychological Science, vol. 16, no. 2, and women difer in their acceptance?” Transportation Re- pp. 315–320, 2008. search Part A: Policy and Practice, vol. 148, pp. 183–198, 2021. [50] W. Qazi, S. A. Raza, and N. Shah, “Acceptance of e-book [34] M. A. Ribeiro, D. Gursoy, and O. H. Chi, “Customer ac- reading among higher education students in a developing ceptance of autonomous vehicles in travel and tourism,” country the modifed difusion innovation theory,” In- Journal of Travel Research, vol. 61, no. 3, pp. 620–636, 2021. ternational Journal of Business Information Systems, vol. 27, [35] Z. B. Ramadan, M. F. Farah, and M. Mrad, “An adapted TPB no. 2, pp. 222–245, 2018. approach to consumers’ acceptance of service-delivery [51] E. M. Rogers, “Difusion of innovation theory,” 1995, https:// drones,” Technology Analysis & Strategic Management, sphweb.bumc.bu.edu/otlt/mph-modules/sb/ vol. 29, no. 7, pp. 817–828, 2016. behavioralchangetheories/behavioralchangetheories4.html. [36] P. Liu, Q. Guo, F. Ren, L. Wang, and Z. Xu, “Willingness to [52] E. M. Rogers, Difusion or Innovations, Free Press, New York, pay for self-driving vehicles: infuences of demographic and NY, USA, 2003. psychological factors,” Transportation Research Part C: [53] Z. Rezvani, J. Jansson, and J. Bodin, “Advances in consumer Emerging Technologies, vol. 100, pp. 306–317, 2019. electric vehicle adoption research: a review and research [37] M. L. Cunningham, M. A. Regan, S. A. Ledger, and agenda,” Transportation Research Part D: Transport and J. M. Bennett, “To buy or not to buy? Predicting willingness to Environment, vol. 34, pp. 122–136, 2015. pay for automated vehicles based on public opinion,” [54] R. Agarwal and J. Prasad, “Te role of innovation charac- Transportation Research Part F: Trafc Psychology and Be- teristics and perceived voluntariness in the acceptance of haviour, vol. 65, pp. 418–438, 2019. information technologies,” Decision Sciences, vol. 28, no. 3, [38] L. M. Hulse, H. Xie, and E. R. Galea, “Perceptions of au- pp. 557–582, 1997. tonomous vehicles: relationships with road users, risk, gender [55] K. B. Ooi, J. J. Sim, K. T. Yew, and B. Lin, “Exploring factors and age,” Safety Science, vol. 102, pp. 1–13, 2018. infuencing consumers’ behavioral intention to adopt [39] R. Shabanpour, N. Golshani, A. Shamshiripour, and broadband in Malaysia,” Computers in Human Behavior, A. K. Mohammadian, “Eliciting preferences for adoption of vol. 27, no. 3, pp. 1168–1178, 2011. fully automated vehicles using best-worst analysis,” Trans- [56] Y. S. Wang, S. C. Wu, H. H. Lin, Y. M. Wang, and T. R. He, portation Research Part C: Emerging Technologies, vol. 93, “Determinants of user adoption of web ‘automatic teller pp. 463–478, 2018. machines’: an integrated model of ’transaction cost theory’ [40] C. A. Spurlock, J. Sears, G. Wong-Parodi et al., “Describing and ’innovation difusion theory,” Service Industries Journal, the users: understanding adoption of and interest in shared, vol. 32, no. 9, pp. 1505–1525, 2012. electrifed, and automated transportation in the San Francisco [57] K. C. Chung, “Gender, culture and determinants of behav- Bay Area,” Transportation Research Part D: Transport and ioural intents to adopt mobile commerce among the Y Environment, vol. 71, pp. 283–301, 2019. Generation in transition economies: evidence from [41] C. Bernhard, D. Oberfeld, C. Hofmann, D. Weismuller, and Kazakhstan,” Behaviour & Information Technology, vol. 33, H. Hecht, “User acceptance of automated public transport,” no. 7, pp. 743–756, 2013. Transportation Research Part F: Trafc Psychology and Be- [58] J. Y. M. Kang, J. M. Mun, and K. K. P. Johnson, “In-store haviour, vol. 70, pp. 109–123, 2020. mobile usage: downloading and usage intention toward [42] K. F. Yuen, L. Cai, G. Qi, and X. Wang, “Factors infuencing autonomous vehicle adoption: an application of the tech- mobile location-based retail apps,” Computers in Human nology acceptance model and innovation difusion theory,” Behavior, vol. 46, pp. 210–217, 2015. Technology Analysis & Strategic Management, vol. 33, no. 5, [59] M. S. Sharifzadeh, C. A. Damalas, and G. Abdollahzadeh, pp. 505–519, 2020. “Predicting adoption of biological control among Iranian rice Journal of Advanced Transportation 17 farmers: an application of the extended technology acceptance [78] National Bureau of Statistics, “Bulletin of the seventh national model (TAM2),” Crop Protection, vol. 96, pp. 88–96, 2017. census (4th),” 2021, http://www.stats.gov.cn/english/ PressRelease/202105/t20210510_1817185.html. [60] S. Min, K. K. F. So, and M. Jeong, “Consumer adoption of the [79] R. R. McCrae, J. E. Kurtz, S. Yamagata, and A. Terracciano, Uber mobile application: insights from difusion of in- “Internal consistency, retest reliability, and their implications novation theory and technology acceptance model,” Journal of for personality scale validity,” Personality and Social Psy- Travel & Tourism Marketing, vol. 36, no. 7, pp. 770–783, 2018. chology Review, vol. 15, no. 1, pp. 28–50, 2011. [61] S. Nordhof, J. de Winter, R. Madigan, N. Merat, B. van Arem, [80] M. Tavakol and R. Dennick, “Making sense of Cronbach’s and R. Happee, “User acceptance of automated shuttles in alpha,” International Journal of Medical Education, vol. 2, Berlin-Schoneberg: ¨ a questionnaire study,” Transportation pp. 53–55, 2011. Research Part F: Trafc Psychology and Behaviour, vol. 58, [81] Congress, “Privacy plan required for highly automated ve- no. 58, pp. 843–854, 2018. hicles,” 2021, https://www.congress.gov/bill/117th-congress/ [62] Y. Duan, Q. He, W. Feng, D. Li, and Z. Fu, “A study on e-learning house-bill/3711/text?r=1. take-up intention from an innovation adoption perspective: a case [82] L. Li, X. He, G. A. Keoleian et al., “Life cycle greenhouse gas in China,” Computers & Education, vol. 55, no. 1, pp. 237–246, emissions for last-mile parcel delivery by automated vehicles and robots,” Environmental Science & Technology, vol. 55, [63] A. A. A. Hamad, I. Petri, Y. Rezgui, and A. Kwan, “Towards no. 16, Article ID 11360, 2021. the innovation of an integrated ‘One-Stop-Shop’ online services utility management: exploring customer’ technology acceptance,” Sustainable Cities and Society, vol. 34, pp. 126– 143, 2017. [64] Rahul, “Autonomous vehicles deliveries vs. Drone deliveries,” 2019, https://blog.route4me.com/autonomous-vehicle- deliveries-vs-drone-deliveries/. [65] Tencent, “Shocking! A delivery man urinated in a customer’s food,” 2021, https://new.qq.com/omn/20211205/ 20211205A02NM000.html. [66] H. P. Young, “Innovation difusion in heterogeneous pop- ulations: contagion, social infuence, and social learning,” Te American Economic Review, vol. 99, no. 5, pp. 1899–1924, [67] C. Higgins, D. Compeau, and D. Meister, “From prediction to explanation: reconceptualizing and extending the perceived characteristics of innovating,” Journal of the Association for Information Systems, vol. 8, no. 8, pp. 409–439, 2007. [68] J. P. Roderigue, Te Geography of Transport System, Rout- ledge, New York, NY, USA, 2020. [69] M. A. Figliozzi, “Carbon emissions reductions in last mile and grocery deliveries utilizing air and ground autonomous ve- hicles,” Transportation Research Part D: Transport and En- vironment, vol. 85, 2020. [70] G. Xi, X. Cao, and F. Zhen, “How does same-day-delivery online shopping reshape social interactions among neighbors in Nanjing?” Cities, vol. 114, Article ID 103219, 2021. [71] X. Li, “White rhino comes into our life, whose delivery ca- pabilities are more than couriers,” Jiading Newspaper, 2021. [72] A. L. Comrey and H. B. Lee, A First Course in Factor Analysis, Erlbaum, Hillsdale, MI, USA, 1992. [73] R. B. Kline, Principles and Practice of Structural Equation Modeling, Guilford Publications, New York, NY, USA, 2016. [74] T. Zhang, D. Tao, X. Qu et al., “Automated vehicle acceptance in China: social infuence and initial trust are key de- terminants,” Transportation Research Part C: Emerging Technologies, vol. 112, pp. 220–233, 2020. [75] C. Fornell and D. F. Larcker, “Evaluating structural equation models with unobservable variables and measurement error,” Journal of Marketing Research, vol. 18, no. 1, pp. 39–50, 1981. [76] J. F. Hair, W. C. Black, and B. J. Babin, Multivariate Data Analysis, Cengage Learning, Boston, MA, USA, 2014. [77] C. M. Cheung, R. P. Zhang, R. Wang, S. C. Hsu, and P. Manu, “Group-level safety climate in the construction industry: infuence of organizational, group, and individual factors,” Journal of Management in Engineering, vol. 38, no. 1, 2022.

Journal

Journal of Advanced Transportation – Hindawi Publishing Corporation

Published: Mar 8, 2023

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

Customer’s Adoption Intentions toward Autonomous Delivery Vehicle Services: Extending DOI Theory with Social Awkwardness and Use Experience

References (73)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies