Kandó Kálmán Doctoral School of Transportation and Vehicle Engineering  

BME KJK, Department of Transport Technology and Economics

Supervisor: Dr. ESZTERGÁR-KISS Domokos

Revealing the Preferences for Mobility as a Service

Introducing the research area

Transportation-related issues (e.g., congestion and greenhouse gas emissions) in cities worldwide are becoming increasingly serious. Alternative and innovative solutions are needed to enable people to travel sustainably. Thanks to the development of digital technologies, shared transport modes are more present in cities to complement public transport (PT) services [1]. However, more transport opportunities often lead to inconvenience for travelers since they need to use several different mobile applications (apps) to plan a trip. Thus, Mobility as a Service (MaaS) is introduced to integrate transport modes equipped with routing, booking, and payment functionalities in one single app (Figure 1). MaaS is often offered as mobility bundles consisting of the usable amount of transport modes. Promoting MaaS by considering travelers’ needs could potentially reduce negative transport externalities.

 

Figure 1. The illustration of MaaS.

Brief introduction of the research place

The research is conducted at the Department of Transport Technology and Economics, KJK BME, under the supervision of Dr. Domokos Esztergár-Kiss. At the department, relevant international conferences were organized (e.g., MT-ITS 2015, EWGT 2017, and hEART 2019) and several Horizon 2020 projects, Interreg projects, and COST Actions have been conducted (e.g., MoveCit, MaaS4EU, Electric traveling, BE OPEN, RegiaMobil). Surveys to gather the data were arranged in Hungary. Moreover, international cooperations were made with TU Delft, the University of Catania, and the Silesian University of Technology to bring a deeper understanding of travelers' preferences for MaaS. Before the general Hungarian population survey, we first distributed a pilot survey to BME students, allowing us to assess the instruments.

History and context of the research [S1,S4]

This research is primarily around travel behavior, where the intersection of transportation engineering, behavioral science, and economics is explored, allowing the creation of novel solutions to urban transportation challenges. A systematic literature review is conducted to identify the research gaps.

MaaS aims to tackle the challenge of adopting new mobility services and maximizing the benefits of the services by introducing a combined framework [2]. Earlier works have examined user preferences toward multimodal journey planners [3,4]. However, MaaS is a relatively new concept, and one technology that needs to be developed is a MaaS app. This raises the question of what functionalities of the MaaS app are demanded by travelers.

The potential users of MaaS need to be identified to maximize the uptake of the service. Studies on preferences for MaaS primarily include several factors, such as socio-demographic, travel [5,6], and attitudinal characteristics [7,8]. In general, early adopters of MaaS are younger, educated individuals with a higher income and those who often use PT. While the general findings on MaaS adopters have been revealed, the heterogeneity in MaaS preferences has yet to be widely explored. There are no solutions that fit all groups of people. Thus, segregating travelers based on their characteristics is beneficial for determining the suitable promotions of MaaS. In addition, the benefits of MaaS could only be materialized if we consider stakeholders’ expectations.

The research goals, open questions

The primary aim of the research is to identify the preferences for MaaS. Based on the literature review and gap analysis, the following research questions are defined:

-       Which functionalities of the MaaS app are preferred by travelers?

-       Which attitudinal factors are the most important in explaining the preferences for MaaS?

-       Who are the potential adopters of MaaS?

-       How are the users’ preferences for MaaS bundles?

-       What are the expectations of stakeholders toward MaaS?

Methods

Figure 2 presents and summarizes the methods of the research. A survey using an online questionnaire was conducted to obtain primary data. Participants were asked to answer several sets of questions, including socio-demographic and travel characteristics, preferences for MaaS aspects, and attitudinal factors. The preferences for MaaS aspects and attitudinal factors were analyzed using the structural equation modeling. Additionally, the attitudinal factors that were hypothesized to have effects on the behavioral intention to adopt MaaS were segmented into two groups based on the travel captivity and the shared mobility usage experience.

Meanwhile, travel characteristics were used to identify the heterogeneity of participants. The latent class cluster analysis (LCCA) was performed to reveal the sub-groups of respondents, where the transport mode usage frequency was treated as an indicator in the modeling. The transport mode usage frequency was measured on a scale from 1 (never) to 8 (several times per day). Socio-demographic and attitudinal factors were included in the model as covariates. The LCCA is a model-based clustering method following a probabilistic approach to assign individuals to classes, where the optimal number of classes is determined using statistical measurements.

Once the travelers’ groups were revealed, the MaaS bundles (i.e., mobility bundles consisting of a certain amount of mobility services, typically offered as subscriptions) were created based on the groups’ characteristics. MaaS bundles represent the integrative and collaborative idea of mobility solutions, supporting the underutilized yet environmentally friendly transport modes (e.g., bike-sharing, e-scooter-sharing) by combining them with more popular modes such as PT. A stated choice survey was conducted, where participants were provided with six randomized hypothetical scenarios in which they were asked to choose between four options (i.e., three predefined mobility bundles and one pay-as-you-go option). The hypothetical scenarios were constructed using the “rotation.design” function from the “support.CEs” realized in R.

 

Figure 2. Methodological framework.

 

The bundle consists of several mobility services that were available in Budapest and add-ons (e.g., gym membership and online shopping voucher). Add-ons were included in mobility bundles as a nudging strategy to make the bundles more attractive. Advanced discrete choice modeling (i.e., hybrid choice modeling) was applied to examine the travelers' preferences for MaaS bundles, respecting the effects of attitudinal factors. The model was estimated using the APOLLO package [9] in R.

To fully understand the preferences for MaaS, we need to examine the multi-stakeholder viewpoints. The objectives and criteria are defined based on earlier literature. MaaS-related stakeholders are provided with pairwise comparison matrices, and the fuzzy analytic hierarchy process (fuzzy-AHP) is applied to obtain the weights of objectives and criteria.

Results

Preferred functionalities of the MaaS app [S2,S5,S9]

Figure 3 illustrates the functionalities of the MaaS app that are demanded by participants, where the font size indicates the preferences (i.e., the preferred functionalities are shown in the bigger font size).

Figure 3. The functionalities of the MaaS app.

 

Travel time is the most demanded function, followed by the number of transfers and real-time information. In the case of booking and payment methods, the application platform and online bank card are the most favorable options. Additionally, the participants have high preferences for QR codes for the validation process. MaaS operators could consider the participants’ most preferred routing aspects to be incorporated in the main part of the routing screen of the app, while other aspects could be placed in the settings.

 

Attitudinal factors behind MaaS preferences [S3,S6]

MaaS is considered a mobility service product leveraged by the development of technology. Thus, the use of a unified theory of acceptance and use of technology (UTAUT) is adequate. The novel methodological framework is developed by introducing several additional and moderator variables (Figure 4). The validity and reliability of the variables are confirmed by assessing the convergent validity and composite reliability.

According to the path coefficient values, effort expectancy is the most influential factor in determining the behavioral intention to adopt MaaS. It indicates that individuals who believe MaaS will be easy to learn are more inclined to use MaaS. The positive coefficient of social impact means that people will use MaaS if the service is praised by others and receives a positive evaluation from the media. Thus, MaaS operators should collaborate with the media in promoting the service. To meet the performance expectancy, MaaS operators should try to minimize the sore points of traveling, such as the number of transfers and transfer time. The integration between modes is also expected (e.g., micro-mobility services are available near main stations) to support shorter trips that are not covered by PT services. Generally, choice users and people who have tried using shared-mobility services have positive standpoints toward MaaS.

 

Figure 4. The framework of attitudinal variables influencing MaaS preferences and path coefficients.

 

Heterogeneity in mode choice [S7,S8,S10]

Based on the Bayesian Information Criterion (BIC) values, the three-cluster solution has the least value; thus, the solution is selected for further analysis. Figure 5 illustrates the usage level of transport modes and the socio-demographic characteristics of the identified clusters. The first cluster is named the PAS cluster due to the highest level of PT usage, Active modes, and Shared mobility services. Meanwhile, the second cluster is labeled Car since it is evident that travelers in this cluster use cars far more often than respondents in other clusters. The third cluster is labeled Less-travel since there are no stand-out values of transport mode usage frequency found in this cluster.

 

 

 

 

Figure 5. The transport mode usage level and socio-demographic characteristics of the identified clusters.

 

The PAS cluster is characterized by younger travelers and high-school graduates. Concerning employment status, around two-thirds of the travelers falling into this cluster are employed individuals. The majority of people in this cluster are females and middle-income earners. Meanwhile, slightly less than two-thirds of males are present in the Car cluster. Unsurprisingly, older, and higher-income individuals mostly belong to this cluster. Most people in this cluster are college graduates. Lastly, the Less-travel cluster is characterized by females with lower education and income levels. A higher share of maternity leavers and pensioners are present in this cluster.

Regarding the willingness to adopt MaaS, the PSA cluster generally has positive standpoints toward MaaS. Individuals in this cluster state that they are willing to adopt MaaS once it is implemented and decrease the use of their cars. People in the Less-cluster mainly show high deterrence against MaaS. Finally, travelers in the Car cluster are generally not willing to give up their cars even when MaaS is available.

Expected impact and further research

The findings and developed methods will be beneficial in creating MaaS ecosystems that suit users’ preferences and in formulating relevant policies to support the wider acceptance of the service. Further research will be focused on the creation of MaaS bundles that respect the users' characteristics [S11]. Furthermore, to gain a complete concept of preferences for MaaS, the viewpoints of MaaS-related actors will be examined. By comprehensively uncovering the preferences for a sustainable mobility service, long-term strategies could be realized, such as promoting intermodality behavior by creating mobility hubs, reliable bike lanes and pedestrians, and low-emission zones. Long-term impacts could be expected, such as improved air quality, more interactions between citizens, encouraging active travel, and eventually improved well-being.

Publications, references, links

List of corresponding own publications.

[S1] Kriswardhana W, Esztergár-Kiss D. A systematic literature review of Mobility as a Service: Examining the socio-technical factors in MaaS adoption and bundling packages. Travel Behaviour and Society 2023;31:232–43. https://doi.org/https://doi.org/10.1016/j.tbs.2022.12.007. (IF: 5.85)

[S2] Kriswardhana W, Esztergár-Kiss D. Exploring the aspects of MaaS adoption based on college students’ preferences. Transport Policy 2023;136:113–25. https://doi.org/https://doi.org/10.1016/j.tranpol.2023.03.018. (IF: 6.183)

[S3] Kriswardhana W, Esztergár-Kiss D. University students’ adoption of mobility as a service with respect to user preferences and group differences. Journal of Public Transportation 2024;26:100079. https://doi.org/https://doi.org/10.1016/j.jpubtr.2023.100079. (IF: 12.2)

[S4] Kriswardhana W, Esztergár-Kiss D. Transport Modes in MaaS Packages and Their Impacts on Transport Externalities: A Literature Review. XVI. IFFK Conference 2022, 2022.

[S5] Kriswardhana W, Esztergar-Kiss D. Cluster Analysis of User Preferences related to MaaS Aspects. 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems (MT ITS) MT-ITS 2023, 2023. https://doi.org/10.1109/MT-ITS56129.2023.10241445.

[S6] Kriswardhana W, Esztergár-Kiss D. College students’ preferences toward Mobility as a Service based on the UTAUT model. XIII. International Conference on Transport Sciences 2023, Győr, Hungary: Széchenyi István University; 2023, p. 59–67.

[S7] Kriswardhana W, Esztergár-Kiss D. Identifying heterogeneity in university students’ transport mode choice. IOP Conference Series: Earth Environmental Science, vol. 1294, IOP Publishing; 2024, p. 12027. https://doi.org/10.1088/1755-1315/1294/1/012027

[S8] Kriswardhana W, Esztergár-Kiss D. Identifying Latent Mobility as a Service Preference Segment among College Students. (under review)

[S9] Kriswardhana W, Esztergár-Kiss D. Examining University Students’ Preferences toward MaaS Aspects. (under review)

[S10] Kriswardhana W, Esztergár-Kiss D. Heterogeneity in Transport Mode Choice of College Students at a University Based on the MaaS Concept. (under review)

[S11] Kriswardhana W, Esztergár-Kiss D. The Role of Intermodality and Environmental Consciousness in the Preferences for MaaS Bundles: A Hybrid Choice Modeling Approach. (under review)

 

Table of links.

https://ktkg.bme.hu/en/

https://scholar.google.com/citations?user=XqBTgY0AAAAJ&hl=en

 

List of references.

[1]  Zuniga-Garcia N, Gurumurthy KM, Yahia CN, Kockelman KM, Machemehl RB. Integrating shared mobility services with public transit in areas of low demand. J Public Transp 2022;24. https://doi.org/10.1016/j.jpubtr.2022.100032.

[2]  Lopez-Carreiro I, Monzon A, Lois D, Lopez-Lambas ME. Are travellers willing to adopt MaaS? Exploring attitudinal and personality factors in the case of Madrid, Spain. Travel Behav Soc 2021;25:246–61. https://doi.org/10.1016/j.tbs.2021.07.011.

[3]  Ferreira MC, Fontesz T, Costa V, Dias TG, Borges JL, E Cunha JF. Evaluation of an integrated mobile payment, route planner, and social network solution for public transport. Transp. Res. Procedia, vol. 24, 2017, p. 189–96. https://doi.org/10.1016/j.trpro.2017.05.107.

[4]  Lopez-Carreiro I, Monzon A, Lopez E, Lopez-Lambas ME. Urban mobility in the digital era: An exploration of travellers’ expectations of MaaS mobile technologies. Technol Soc 2020;63. https://doi.org/10.1016/j.techsoc.2020.101392.

[5]  Zijlstra T, Durand A, Hoogendoorn-Lanser S, Harms L. Early adopters of Mobility-as-a-Service in the Netherlands. Transp Policy 2020;97:197–209. https://doi.org/10.1016/j.tranpol.2020.07.019.

[6]  Hoerler R, Stünzi A, Patt A, Del Duce A. What are the factors and needs promoting mobility-as-a-service? Findings from the Swiss Household Energy Demand Survey (SHEDS). Eur Transp Res Rev 2020;12. https://doi.org/10.1186/s12544-020-00412-y.

[7]  Schikofsky J, Dannewald T, Kowald M. Exploring motivational mechanisms behind the intention to adopt mobility as a service (MaaS): Insights from Germany. Transp Res Part A Policy Pract 2020;131:296–312. https://doi.org/10.1016/j.tra.2019.09.022.

[8]  Ye J, Zheng J, Yi F. A study on users’ willingness to accept mobility as a service based on UTAUT model. Technol Forecast Soc Change 2020;157. https://doi.org/10.1016/j.techfore.2020.120066.

[9]        Hess S, Palma D. Apollo: A flexible, powerful and customisable freeware package for choice model estimation and application. J Choice Model 2019;32:100170. https://doi.org/10.1016/J.JOCM.2019.100170.