Keywords

1 Introduction

As a new model of transportation, car sharing services are becoming an important mode of transportation. The shortage of traffic resources and road resources that cannot be changed in a short period of time determines that car sharing service has great development potential and strong vitality. As a new mode, car-sharing has aroused the interest of youth group while China’s car sharing services are still in the primary stage of market operation. Many enterprises have explored the mode of car-sharing, but no mature solution has been developed for the user experience problems in operation. Issues of the user experience in car-sharing services still need to be resolved to enhance user’s acceptance and satisfaction of shared vehicles.

The objective, based upon interview and the Decision-Making Trial and Evaluation Laboratory (DEMATEL) analysis, is to identify critical factors influencing the user experience in car sharing services, as well as to assess the interrelationships among the main influencing factors, which will eventually serve as a novel tool model for future user experience improvement. This study aims to provide valuable references for the car sharing platforms to improve the user experience quality, clarify the design and development direction of related products and services, and advance the healthy development of the car sharing model.

2 Literature Review

The literature research is carried out through databases and search engines, primarily initiated through ScienceDirect, Web of Science, EBSCOhost, CNKI and Google Scholar but includes findings from papers reference sections as well. The research started from collecting studies from database, based on 6 keywords (“car sharing”, “shared car”, “shared vehicle”, “user experience”, “service quality” and “user satisfaction”), for the publication year range 2011 to 2019. The authors keep the relevant literatures according to the abstracts, and extract the relevant factors that affect the user experience of the car sharing service from the articles.

Lamberton et al. [1] proposed that the influence of price on transportation preferences equally applies to car sharing services. Joshua Paundra et al. [2] carried out that the three instrumental attributes, price, parking convenience and car type, influence people’s intention to select a shared car. Stefano de Luca and Roberta Di Pace [3] found that the price increase of a car sharing service reduces the willingness to switch from private car to a car sharing service. Kaspi et al. [4] found that car sharing services are more attractive to potential users, if parking places are included in the service. Lower price, more convenience parking, and cars of the electric type linked to higher intention to select a shared car. Alexandre de Lorimier et al. [5] found that the combination of high availability and high vehicle usage ensures users satisfaction with the service and that vehicles are used. The number of vehicles parked at a station has the most effect on availability, and vehicle usage is affected by average vehicle age, and by member concentration in the vicinity of stations. According to the research of Cartenì et al. [6], users are sometimes attracted to carsharing service because of its positive environmental image. Stefan Illgen and Michael Höck [7] found that key success factors of a good car sharing service included advantageous relations between the market environment (e.g. electricity and fuel prices) and important characteristics of electric cars (e.g. price and range). Guido Perboli et al. [8] compared the business models of the four car sharing companies, and analyzed the advantages and disadvantages of each model. If car sharing and other shared services are to gain in popularity, there must be ample trust or reputation [9, 10]. Vehicle availability, brand reputation, price model, travel flexibility, charging convenience, station distribution, sharing model and web & app fluency are the influencing factors of the user experience that can be extracted from their research. Through depth interview with Sydney’s car sharing users, Robyn Dowling et al. [11] explores the factors in car sharing, and divides them into three categories: Digital attributes (APP and web page, online payment, reservation system, etc.), material attributes (car cabin environment, parking space, car condition and performance, etc.) and social attributes (in car passengers, travel flexibility, entertainment, etc.). Xusen Cheng et al. [12] found factors that influence online and offline car sharing service quality, including structure assurance (safety, comfort, safe transaction environment), platform responsiveness (response speed and quality), information congruity (service conforms to the content displayed on the mobile platform) and reservation service. Min Qu et al. [13] developed an improved approach to evaluate car sharing options, and proposed 24 indicators from four dimensions: economic, environmental, social, and car sharing system performance. The influencing factors of user experience include mobility, accessibility, security, functional diversity, quality of service, accessibility for mobility impaired groups, etc. Jinhee Kim et al. [14] found that satisfaction significantly affects the car-sharing decision, and that car availability has a significant effect on the likelihood of joining a car-sharing organization. Francesco Ferrero et al. [15] mentioned that mode identifies the different ways in which a car-sharing service can be provided. Station-based mode [16] and the Free-floating mode [17] affects the flexibility of the shared car travel.

3 Method

3.1 DEMATEL Procedure

Decision-Making Trial and Evaluation Laboratory (DEMATEL) method is often used to analyze the complex relationship between management problems. It is a methodology put forward by Battelle Geneva Research Center in 1971 to solve the complex and difficult problems between science, technology and human beings in the real world and clarify the essence of the problems [18]. This approach has been widely used in many fields such as smart product service system [19], hospital performance management [20], sustainable online consumption [21] and logistics 4.0 [22]. As a type of structural modeling method, DEMATEL method can simplify the complex problems and prompt the logical relationship between the factors through determinant calculation [23, 24]. The method not only converts the interdependencies of factors into cause-and-effect relationships but also determines the critical factors of a system aided by impact relation diagrams [25].

The DEMATEL procedure in this study is carried out in the following steps: (1) identifying the main influencing factors that impact the satisfaction of user experience in the car sharing service; (2) collecting opinions of target users to estimate the influence of factors on each other using the DEMATEL questionnaire with one of four level values; (3) generating the direct relation matrix Z; (4) Calculating the λ value, the normalized direct relation matrix and the direct/indirect relation matrix T, and the corresponding Prominence value (D + R), and Relation value (D − R) of each influencing factor; (5) obtaining the influence relation map (IRM) with (D + R) value and (D − R) value of each influencing factor to help observe the interrelationship structure; (6) extracting the key influencing factors from the 14 factors to identify the relationship among all factors. Based on the previous research, this research explored the influence, causality and importance of the factors that affect the user experience in car sharing services.

3.2 Fourteen Main Influencing Factors

This study applies literature research and user research methods (questionnaire, observation and in-depth interview) to mine the influencing factors of user experience in car sharing service. A large number of target users were investigated to collect users’ opinions toward the car sharing travel. The authors analyzed the information collected from related literatures and user research, to extract the factors that have impact on their perception of the satisfaction with car sharing service.

Users are divided into four groups according to the literature review and user research, including Loyal Sharing type, Restriction Alternative type, Business Travel type and Free-outing type. The Loyal Sharing group refers to the young people who have driver’s license but have no car in the city, and use the sharing car to meet the rigid demand of travel. The Restriction Alternative group refers to the people whose private cars are restricted to drive on working days, and the shared cars are used to solve the restriction problem. The Business Travel group refers to the business travel group with high requirements for privacy, which needs to avoid the interference of noisy environment. The Free-outing group refers to young people/families who pay attention to the freedom of travel and people who try to use shared cars because of curiosity and freshness.

Different user groups have different emphasis on the evaluation of user experience. After analyzing the information collected by the above methods, the influencing factors categories of different user groups are sorted out. And as shown in Table 1, the following 21 preliminary influencing factors of the user experience in car sharing service are summarized and defined.

Table 1. 22 preliminary user experience influencing factors.
Full size table

The 22 influencing factors are clustered by using card sorting method, and finally 14 main influencing factors are sorted out and given serial number Factor 1 to Factor 14 as follows: (1) Factor 1 - Vehicle Condition and Performance; (2) Factor 2 - Available Supporting Service Facilities; (3) Factor 3 - Safety of Vehicle Service; (4) Factor 4 - Pick Up & Return Mode; (5) Factor 5 - Price Model; (6) Factor 6 - Brand and Visual Image; (7) Factor 7 - Simple Learning & Using; (8) Factor 8 - Vehicle Efficiency; (9) Factor 9 - Traffic Transferring Convenience; (10) Factor 10 - Brand & Model Type Diversity; (11) Factor 11 - Environment Cleanliness; (12) Factor 12 - Real-time Feedback; (13) Factor 13 - Pleasure and Novelty; and (14) Factor 14 - Value Identification.

These 14 influencing factors form the items of DEMATEL questionnaire in this study, and the specific definitions of each influencing factor of car sharing service are shown in Table 2.

Table 2. Specific definitions of the 14 influencing factors of user experience.
Full size table

3.3 DEMATEL Questionnaire

This study focuses on the general young users of car sharing service, and uses the 14 user experience factors summarized in the previous section to make the DEMATEL questionnaire. Expert interview and questionnaire were used in the process of investigation. The questionnaire was mainly distributed to the young people who have shared car use experience in the first and second tier cities.

The DEMATEL questionnaire (Table 3) is used for shared car users to estimate the direction of interaction and the degree of relative priority of each factor listed in first column to each factor listed in first row in Table 3. The priority degrees of factors in first column over factors in first row are defined in 0–4 five levels, i.e., value ‘0’ means “no impact”, value ‘1’ means “low impact”, value ‘2’ means “distinct impact”, value ‘3’ means “big impact”, and value ‘4’ means “extreme impact”.

Table 3. The DEMATEL questionnaire.
Full size table

Due to the poor readability of the form questionnaire, we transform the DEMATEL questionnaire into a normal questionnaire which is easy for users to understand and fill in for data collection. A total of 58 questionnaires were collected from May 29 to June 6, 2019, including 51 valid ones. And each questionnaire is filled in 15–40 min.

4 Data Analysis and Results

4.1 Direct Influence Matrix Z

51 pieces of effective data were calculated based on the following equation.

$$ \varvec{Z} = \frac{1}{n}\sum\nolimits_{m = 1}^{n} {\left[ {z_{ij}^{m} } \right],\quad i,j = 1,2,3, \ldots ,k.} $$
(1)

The elements in direct influence matrix Z of the interactions between 14 main factors are obtained as listed in Table 4.

Table 4. The direct influence matrix Z.
Full size table

4.2 Matrix of Total Relations T

Users only estimated the direct influence of factors on each other. Hence, the total influence matrix T is obtained from matrix X by applying the transition theory and summing up all direct and indirect effects. The matrix of total relations T (Table 5) was derived, by using Eqs. (2)–(4). Matrix X is the normalized direct relation matrix, and I is the identity matrix.

Table 5. Matrix of total relations T.
Full size table
$$ \varvec{X} = \left[ {x_{ij} } \right]_{k \times k} = s\varvec{Z}, $$
(2)
$$ s = min\left( {\frac{1}{{\mathop {\hbox{max} }\limits_{1 \le j \le k} \sum\nolimits_{i = 1}^{k} {z_{ij} } }},\frac{1}{{\mathop {\hbox{max} }\limits_{1 \le i \le k} \sum\nolimits_{j = 1}^{k} {z_{ij} } }}} \right), $$
(3)
$$ \varvec{T} = \mathop {\lim }\limits_{m \to \infty } \left( {\varvec{X} + \varvec{X}^{2} + \ldots + \varvec{X}^{m} } \right) = \varvec{X}\left( {\varvec{I} - \varvec{X}} \right)^{ - 1} . $$
(4)

The upper quartile Q3 of all the elements in the total relations matrix T (keep 3 decimal places), i.e., 0.580, is taken as the threshold to measure the strength of interactions between factors. If all values in the row and the column that correspond to a factor in the matrix T are below the threshold value, this factor and the corresponding row and column will be removed.

Then, the upper third of Q3 (0.617) is taken as the boundary value of strong influence relation and general influence relation. Values higher than 0.617 are considered as strong influences, and values between 0.617 and 0.580 are considered as general influences. As shown in Table 6, the general influences are marked with underlines, and the strong influences are marked with bold type and underlines.

Table 6. Matrix of total relations T.
Full size table

4.3 Prominence and Net Effect

Value D and R are achieved by summing up the rows and columns of total relation matrix T, using Eqs. (5)–(6).

$$ \varvec{D}_{i} = \sum\nolimits_{j = 1}^{n} {t_{ij} ,\left( {i = 1,2,3, \ldots ,n} \right)} $$
(5)
$$ \varvec{R}_{i} = \sum\nolimits_{j = 1}^{n} {t_{ij} ,\left( {i = 1,2,3, \ldots ,n} \right)} $$
(6)

Value D is called the degree of influential impact, and value C is called the degree of influenced impact. \( \varvec{D}_{i} \) is the sum of the ith row of matrix T and represents all the direct and indirect effects which are dispatched from Factor i to other factors. And \( \varvec{R}_{j} \) is the sum of the jth column of matrix T and represents all the direct and indirect effects that Factor j receives from the other factors.

The prominence value (D + R) is obtained by adding the influence degree (D) and the influenced degree (R) of the factor.

The prominence value (importance) played by the factor in a system is represented by (D + R) value, where (D − R) value stands for the net effect (i.e. the degree of importance) of the factor contributed to the system. The factors are categorized into two groups of cause (driver) and effect (receiver) factors. If the (D − R) value of a factor is positive, this factor is grouped under the category of driver factors which has an influence on other factors; if the value of (D − R) is negative, this factor is grouped under the category of receiver factors which receives influence from other factors.

The value D, R, (D + R), (D − R) of Factor 1 to Factor 14 were listed in Table 7.

Table 7. Prominence (D + R) and Net effect (D − R) values.
Full size table

The average value of (D + R) of each factor is 14.997 (three decimal places are reserved). The (D + R) value of nine factors is higher than the average (D + R) value, so the importance of factors affecting the user experience in car sharing service is ranked as follows: Factor 8 - Vehicle Efficiency, Factor 2 - Available Supporting Service Facilities, Factor 5 - Price Model, Factor 1 - Vehicle Condition and Performance, Factor 3 - Safety of Vehicle Service, Factor 7 - Simple Learning & Using, Factor 12 - Real-time Feedback, Factor 4 - Pick Up & Return Mode, Factor 6 - Brand and Visual Image.

From the (D − R) ranking, we find that Factor 1 (Vehicle Condition and Performance), Factor 2 (Available Supporting Service Facilities) and Factor 3 (Safety of Vehicle Service) are the three important factors that directly influence other factors; and Factor 14 (Value Identification), Factor 6 (Brand and Visual Image) and Factor 13 (Pleasure and Novelty) are the three factors that are most influenced by other factors.

4.4 The Influential Relation Diagram

The influential relation diagram can be plotted in the (\( D_{i} + R_{i } , D_{i} - R_{i} \)) layout, using \( \left( {D_{i} + R_{i } } \right) \) as the horizontal axis and (\( D_{i} - R_{i} \)) as the vertical axis. Fourteen user experience influencing factors of car sharing services are marked in the coordinate axis, and the influential relation diagram is made. According to the position of each factor in the diagram, the ones which have a great effect on other factors or have a complicated relationship with other criteria can be found.

The relationship with the influence strength lower than 0.580 will not be marked. The relationship with the influence strength between 0.580 and 0.617 will be regarded as general influence and marked with dotted line; the relationship with the influence strength greater than or equal to 0.617 will be regarded as strong influence and marked with solid line. The specific location of each factor is shown in Fig. 1.

Fig. 1.
figure 1

The influential relation diagram of 14 influencing factors of user experience in car sharing service.

Full size image

As can be seen from Fig. 1, Factor 1 (Vehicle Condition and Performance), Factor 2 (Available Supporting Service Facilities) and Factor 3 (Safety of Vehicle Service) are the most important influencing factors in the car sharing travel service. Factor 14 (Value Identification), Factor 6 (Brand and Visual Image) and Factor 13 (Pleasure and Novelty) are the three most directly/indirectly affected factors in the car sharing experience. Factor 8 (Vehicle Efficiency) and Factor 5 (Price Model) are two important influencing factors in the shared car service.

5 Discussion and Conclusion

5.1 Key Factors Influencing the User Experience in Car Sharing Service

The first three critical factors of car sharing user experience, the first three influencing factors and the influenced factors of the car sharing user experience are summarized in Table 8.

Table 8. Key influencing factors of user experience in car sharing service.
Full size table

It is found that the first three key influencing factors of user experience in car sharing services are Factor 8 (Vehicle Efficiency), Factor 2 (Available Supporting Service Facilities) and Factor 5 (Price Model). The results show that users pay most attention to the efficiency in the process of car reservation, rental and return when using the shared car; secondly, the key influencing factors of user experience are whether the basic service facilities supporting the shared cars, such as charging piles and parking spaces, are perfect, easy to find and use. At the same time, the rental fee and price model of the car sharing service also have an important impact on the user experience.

5.2 Influential Relationships Between Main Factors

Results of the ranking of (D − R) values shows that Vehicle Condition and Performance (Factor 1), Available Supporting Service Facilities (Factor 2) and Safety of Vehicle Service (Factor 3) are the top three factors with the largest positive Net Effect values, implying that these three factors have the strong impact on other factors. The results demonstrated that the basic driving functions and maneuverability of the vehicle itself are the most important factors affecting the user experience of shared vehicle, which are the most basic factors for creating a good experience that directly affects the user’s evaluation of the whole experience. This analysis found evidence that 1) users attach great importance to the vehicle efficiency, so good car sharing service needs to start from each node of the use process to improve efficiency; 2) the perfection and usability of charging piles, parking spaces and other supporting facilities are essential for user experience; and 3) the safety of vehicle driving and mobile payment also has an important impact on the car sharing experience.

According to the analysis of the affected factors, Value Identification (Factor 14), Brand and Visual Image (Factor 6) and Pleasure and Novelty (Factor 13) are the ones being influenced most greatly.

5.3 Analysis and Discussion on Specific Factors

The essence of Factor 1 (Vehicle Condition and Performance), Factor 2 (Available Supporting Service Facilities) and Factor 3 (Safety of Vehicle Service) is the basic needs of shared car users. In the influential relations diagram, we can see that these three factors are not affected by other factors, but mainly have strong impacts on other factors, and they are completely dominant influencing factors of user experience.

As the most intuitive experience content, the usage functions of vehicles are the basis of a good car sharing travel, which have the most direct impact on the evaluation of the whole experience. Therefore, when improving the user experience of shared cars, we should first ensure the improvement of basic functions and facilities. Only when these three factors are satisfied can the optimization of other factors be effective. On the premise of meeting these three factors, the optimization of other factors can achieve better results.

Factor 8 (Vehicle Efficiency) is located on the right-most side of the graph, which shows that it is strongly influenced by four factors (F1, F2, F3 and F12), and influenced by three factors (F4, F7 and F9) in general. Factor 8 also affects four factors (F5, F6, F13 and F14). The (D + R) value and (D − R) value of vehicle efficiency are both high, indicating that Factor 8 has strong influential relationships with other factors. Therefore, Factor 8 is a critical factor of user experience in car sharing service.

Vehicle efficiency is closely related to user experience. Most shared car users are urban youth, and most of the use scenarios are commuting, party and outing on weekdays, so there is a high demand for vehicle efficiency. Efficient car sharing service needs to be built from each node of the use process: for example, simplifying car rental procedures and operations, providing more renting entrances to enable users to reserve vehicles faster; increasing the number of outlets and providing beginning/ending navigation to enable users to get to the parking location faster; improving the features of shared vehicles, enabling users to find the reserved vehicles faster; optimizing the accident-handling process to reduce the time required to remove obstacles. It could improve the efficiency of car sharing service by shortening the time required by each node, which is conducive to improving users’ satisfaction of the whole travel.

Factor 5 (Price Model) is strongly influenced by three basic factors (F1, F2 and F3), and generally influenced by four factors (F4, F8, F9 and F12). At the same time, it affects Factors 6 and Factor 14, and has a lot of mutual interrelationships with other factors, so it is also an important factor affecting user experience and their satisfaction.

Rental costs and price model of car sharing service will affect users’ perception of a car sharing service. The results show that users will mainly evaluate the vehicle price charging mode according to the vehicle condition and performance, supporting service facilities and vehicle service safety; then consider the service quality of the platform and the rapidity and convenience of the service system, and judge whether the price is equivalent to the service. At the same time, the pricing of shared car service will affect the user’s feeling of shared car brand image. Shared car users can be divided into price-sensitive users and quality-oriented users, pursuing high performance-price ratio and high quality respectively. Car sharing platforms should fully consider their brand positioning and target groups when pricing, to formulate a reasonable pricing model.

The interrelationship between Factors 10 (Brand & Model Type Diversity) and other factors is weak. It is only affected by Factors 2 (Available Supporting Service Facilities). Thus, it can be concluded that the contribution value of providing diversified brands and models for users is relatively low in improving the user experience of the whole service system. Therefore, the priority of brand and model diversity is lower than the previous factors when optimizing the car sharing service.

The four factors, Real-time Feedback (Factor 12), Traffic Transferring Convenience (Factor 9), Pick-up & Return Mode (Factor 4) and Simple Learning & Using (Factor 7), are concentrated in the middle of the relationship diagram. These factors do not have influence on each other, and their interrelationships with other factors are mostly general influences. On the premise of guaranteeing the availability factors in car sharing service, optimizing and upgrading the above factors (Performance Quality) will also enhance the user satisfaction accordingly. The only strong influence among the four factors is that Factor 12 (Real-time feedback) significantly affects Factor 8 (Vehicle Efficiency). It can be explained that the real-time feedback of mobile APP (such as vehicle status, location information, order progress information, etc.) can inform users the current status of service, interact with users and give users a sense of control, which plays an important role in improving user experience.

Value Identification (Factor 14) and Pleasure and Novelty (Factor 13) refer to the interest, emotion and value connection established between shared vehicles and users. From the perspective of user demand in Maslow’s demand hierarchy theory, these factors belong to the upper level demand. Users agree with the social values of low-carbon and environmental protection of car sharing, and participate in sharing activities to gain a sense of self-achievement, to achieve self-worth recognition. In the process of using the shared car, through interesting user interface/interaction and the creative facilities, users will have the sense of pleasure, satisfaction and freshness. It makes users have the will to use continuously. In the process of the gradual improvement of the car sharing system, it should also start from establishing emotional relationship with users to enhance good experience and user viscosity.

Factor 11 (Environment Cleanliness) has no significant interrelationship with other factors. However, according to the interview results, the cleaning of the car interior and car body has an important impact on the user satisfaction. It is a visual and olfactory factor that can be directly observed. Therefore, although the interaction between environmental cleanliness and other factors is weak, it is still important to ensure the regular cleaning of car interior and car body in a good car sharing service.

6 Conclusion

This study uses DEMATEL method to analyze the impact of 14 influencing factors on the sharing car user experience. Vehicle condition and performance, available supporting service facilities and safety of vehicle service are the basis of the car sharing service system. As the most intuitive experience content, they have the most direct impact on the evaluation of the whole experience. Good car sharing service should first ensure the availability of vehicles.

Vehicle efficiency is an important key influencing factor of the user experience in the car sharing service. Therefore, all aspects of the process should be designed to improve the efficiency of the whole process. Price model is also one of the most important influencing factors. The price model should conform to the brand positioning, and need to develop reasonable rental mode according to different user groups.

Combining the influence relationships and experience level among the influencing factors in the influential relations diagram, the 14 factors affecting user experience can be divided into three layers (Able-to-use, Easy-to-use and Enjoyable-to-use) to set up the user experience optimization model of car sharing experience. The influence of factors on user experience decreases from bottom to top, and gradually transits from functional factors to emotional factors. Car sharing service system can be optimized by referring to this model: 1) Completing the usability construction of basic functions and service facilities; 2) Designing the usability aspects of the sharing system to simplify the user’s operation and improve efficiency; 3) Establishing an emotional connection with users to make them enjoy the using process and even achieve self-worth.

Fig. 2.
figure 2

The user experience optimization model of car sharing experience.

Full size image

In conclusion, this research provides references and proposals for the user experience design of car sharing service from specific factors, and advances the healthy development of the car sharing model. However, there are limitations in this study. The sample size in this study is small, so in the future research, the research scope should be expanded to make the research more rigorous. In addition, due to the excessive number of questions, respondents may lose patience when filling in questionnaires or participating in the interviews, which may affect the authenticity of the results. In the future, researchers should explore a simpler form of questionnaire to make the results more accurate.