Mobility as a Service (MaaS): How can emerging transport technologies improve mobility inclusiveness?

Under the guidance of Prof. Yoram Shiftan and

Associate Prof. Carlos Lima Azevedo

Ph.D. student- Noam Katzir

Civil And Environmental School Engineering, Technion

​

Introduction

​

​

​

​

​

​

Boxes 5

Technological Innovation in Smart Transportation and Smart Cities

• It examines the ethical, systemic, environmental, and economic impacts that these innovations have on society and the environment.

Yigitcanlar et al., 2020; Vij et al., 2020.

Inclusivity and Vulnerable Social Groups in MaaS

• How MaaS can be designed to be inclusive for vulnerable social groups (VSGs) such as the elderly, individuals with low income, and those with limited technological proficiency.

Lucas, 2019; Dadashzadeh et al., 2022; Hensher et al., 2021.

Subscription-Based Models and Business Approaches in MaaS

• How do these models influence consumer behavior and mobility patterns?

Caiati et al., 2020; Kamargianni & Matyas, 2017.

Research Areas

Research Objectives

To gather insights on making Mobility as a Service (MaaS) more inclusive by increasing its usability for different social and vulnerable groups.

Investigating User Preferences:

Recommending on Business Models:

Comparing Different Cultures

Understanding User Needs and Capabilities:

Explore how inclusiveness affects the willingness to adopt MaaS.

​

Assess preferences for fixed bundles, flexible options, or PAYG.

Identify needs, abilities, adoption behaviors, and other factors influencing MaaS usage.

​

​

Study MaaS adoption across different geographical locations, demographics, and urban environments.

​

​

​

​

​

​

Develop recommendation on how MaaS business models can better support the inclusiveness of various vulnerable groups

Theoretical Background

Literature Review

​

​

​

​

​

​

The Concept of MaaS

​

The Willingness to adopt MaaS

Theories of Justice and Equity

MaaS Inclusiveness

Methodology Review

Theoretical Background- the concept of MaaS

​

​

​

​

​

​

​

Goals:

• Convenience & Flexibility: "One-Stop-Shop" for mobility.
• Sustainability: Reduces car dependency, and improves air quality (Sochor et al., 2016, 2018).
• Broader Societal Goals: Enhances urban livability and supports sustainability and urban planning (Panou et al., 2015; Ho & Hensher, 2020).

​

​

Challenges

• Designing subscription bundles to meet diverse user preferences (Panou et al., 2015; Hensher et al., 2020).

.

Theoretical Background- the concept of MaaS

​

​

​

​

​

​

Willingness to Adopt MaaS

​

• Influencing Factors:
• Service Variables: Time, cost, reliability (Kriswardhana & Esztergár-Kiss, 2023).
• Socio-demographics: Higher education and income levels correlate with greater willingness to use MaaS (Zijlstra et al., 2020).

​

• Barriers:
• Privacy Concerns: User data collection poses acceptance challenges (Cottrill et al., 2020).
• Digital Divide: Risk of exclusion, especially for older individuals (Shirgaokar, 2018).

​

Theoretical Background- Theories of Justice and Equity

​

​

​

​

​

​

Equity

Capability

• Peter Goodwin (1974): Highlighted disparities in the value of time among population groups.
• Digital Access: Inclusion with smart mobility systems (Ruiz-Pérez et al., 2023); Affordability and technological barriers (Smith et al., 2020).

​

​

Equity in transportation

Accessibility

• Ability to reach desired services and activities (Litman, 2021). Interaction between transport, land use, and activity locations (Ben-Akiva & Bowman, 1998).
• Activity Based Accessibility (Nahmias–Biran & Shiftan, 2016).

Accessibility

The Capability Approach

• Amartya Sen's Critique: Shifts focus from primary goods to human capabilities (Sen, 1979, 2005)
• Basic Capability Equality (Nussbaum, 2011).
• Provide necessary resources for all individuals to lead fulfilling lives

MaaS inclusiveness

Framework Overview:

• Evaluation criteria: Accessibility Transport Index (ATI), Accessibility Data Index (ADI), Accessibility Platform Index (API) (Dadashzadeh et al., 2022).
• Developed from the idea of examining the readiness of cities for MaaS.
• Designed for assessing new services or products in the MaaS ecosystem.
• Real-world application for practical evaluation.

​

Key findings:

• Importance of addressing the needs of Vulnerable Social Groups (VSGs).
• Policy recommendations for promoting equity and affordability.
• Role of technology in enhancing inclusivity.

​

Methodology Review

• Discrete choice models, utilizing RP and SP data, are commonly used in transportation research to understand travel behavior )Shiftan et al., 2021).
• Bahamonde-Birke et al. (2023) used Hybrid Choice Models (HCM) in MaaS contexts to address challenges in SP experiments.

• Matyas and Kamargianni (2019) present a survey design incorporating an SP experiment to capture the decision-making process in purchasing MaaS products.

TAM framework explores factors influencing the public's readiness to adopt new technology, encompassing subjective norms, lifestyle habits, and perceived ease of use (Venkatesh & Davis, 2000).

• Qualitative studies, including interviews and focus groups, offer exploration beyond predefined options in transportation (Matyas, 2020; Villeneuve and Kaufmann, 2020),

SP Surveys and Discrete

Choice Modeling

Multi-featured

Products and Service

Technology Acceptance

Model (TAM)

Qualitative approach

Research gaps

​

• The literature highlights various obstacles to transportation faced by individuals with limited access, (Xi et al., 2024). However, there is a notable gap in research regarding how MaaS initiatives can address these barriers for different groups.

​

• There is a scarcity of literature on the application of TAM to analyze the acceptance of MaaS among vulnerable groups.

​

• Our study aims to fill this gap by exploring how integrated mobility services can be customized to meet the transportation needs of different population groups, like the elderly.

​

• Moreover, we have discovered a lack of literature that integrates diverse factors affecting the appeal and uptake of MaaS.

Methodology

General Overview Of Methodology

Objective: Gain comprehensive insights into MaaS adoption and inclusivity. +

Exploring MaaS inclusiveness from the user's viewpoint adds a crucial dimension.

​

​

Approach: Combines qualitative and quantitative research methods.

​

Main Steps:

Theoretical frameworks

Data collection

Survey design

Model estimation

Regulatory analysis

Stakeholder engagement

Practical recommendations

Conceptual Framework

Drawing inspiration from TAM , our framework aims to unravel the User-Centric MaaS Inclusivity and usage behavior of individuals toward MaaS.

Data Collection

• Target Group Interviews/Focus Groups: the elderly, low-income individuals, people with visual impairments, and their caregivers,etc.
• Why we focus on the elderly?

​

• Understand unique requirements, mobility limitations, technology comfort levels, and specific travel challenges.
• Gather qualitative insights to inform the development of our SP, ensuring it captures nuanced preferences and concerns of potential MaaS users.

​

• Stakeholder Engagement: interviews with key stakeholders: policymakers, and transport providers.
• Gather insights crucial for the regulatory analysis stage.
• Provide a holistic understanding of the factors shaping MaaS policies.

​

• Design SP Survey
• Focus on vulnerable groups with a preference for a global comparison.
• Emphasize legal and ethical issues related to MaaS inclusiveness, menu-based subscriptions, or pre-designed bundles based on TAM.

Developing and Estimating a Hybrid Choice Model (HCM)

Developing and estimating an hybrid choice model (HCM) to assess MaaS adoption and acceptance across user segments.

​

​

Log-Sum in Accessibility: The log-sum term is a key component in the calculation of accessibility within discrete choice models. It represents the expected utility an individual can achieve from a set of alternatives. In the context of accessibility, it aggregates the utilities of all potential destinations, providing a comprehensive measure of the accessibility level.

​

The logsum is individual-based measure - capturing the unique utility each person derives from available transportation alternatives

​

Higher log-sum values indicate greater accessibility

​

V ij​ is the systematic utility of alternative 𝑗j for individual 𝑖i. The exponential function 𝑒𝑉𝑖𝑗e transforms the utilities into positive values.

Summing these values across all alternatives and taking the natural logarithm provides the log-sum.

​

This individual-based perspective is crucial for designing MaaS solutions that are tailored to the specific requirements of diverse populations

​

​

​

​

​

Regulatory Sandbox

Objective: A regulatory sandbox enables entrepreneurial development and informs regulatory policy to test their technologies and products in a controlled, supervised environment, mitigating risks and unintended consequences through trial-and-error testing.

​

The UK’s Financial Conduct Authority (FCA) pioneered the concept, which has been adopted worldwide.

​

• Key Activities:
• Testing Measures: Stakeholders test measures within the sandbox.
• Feedback and Discussion: Focus on enhancing inclusivity and sustainability.
• Outcomes:
• Guiding Recommendations: Guide regulatory policies.
• Alignment with Goals: Ensure policies support MaaS goals.
• Regulatory Sandboxes Guide (Jeník & Duff, 2020):
• Objective Setting: Evaluate the necessity and suitability of the sandbox.
• Design: Customize to local conditions using a diligence framework.
• Implementation: Governance, market engagement, testing, and exploring alternatives.
• Focus of Our Sandbox:
• Inclusiveness: Impact of business models on inclusiveness in MaaS.
• Public Role: Promote equitable access through public entities.

Completed Work

Objectives of the Initial Study:

1. Utilize the SP survey to gather some initial insights from potential users regarding their perceptions of MaaS.

​

• Draw preliminary conclusions on the concept of MaaS inclusiveness from the user perspective.

​

3. Use these findings to further develop the initial framework presented in the methodology section.

Research Approach:

​

Methodology:

• Quantitative Approach:
• Utilized a multinomial logit model to analyze transportation preferences.

​

Variables:

• Dependent Variable: Choice of MaaS package. Compare the utilities of MaaS packages to the utility of staying with the current mode of transportation (Reference Point: Situation 4).
• Independent Variables: Survey responses on travel patterns, perceived ease of use, usefulness, privacy concerns, regulation, and package attributes.

​

Utility Functions:

• Linear combinations of parameters and variables for each MaaS package.
• Example:

​

VPACK1​=ASC_pack1+βcar_sharing​×carsharing_pack1+βminibus​×minibus_pack1+…

​

• The same approach was used to define the utility functions for pack2 and pack3.

​

Example for SP Scenario:

Example for SP Scenario:

• As the package number increased, the number of services offered increased accordingly.
• At the end of the survey respondents needed to declare if they prefer in the future to build a personal package or if they prefer a re-designed package.

​

​

Results:

Results:

Insights from the results:

​

Model Comparison:

• Inclusion of Constants:
• Improved likelihood: -1498.08 to -1476.81

​

​

Primary Variables:

• Model 1 [NO CONSTANTS]: Price of MaaS Package: Coefficient = -0.0006
• Use of Public Transportation (PT): Coefficient = 0.019

Willingness to Pay (WTP) Calculation:

​

• Interpretation: Illogical result-> On average, individuals are willing to pay approximately 31.67 NIS for a one-unit increase in public transportation.

​

• Model 2: Price of MaaS Package: Coefficient = -0.001 Use of Public Transportation (PT): Coefficient = 0.0227

​

• Interpretation: On average, individuals are willing to pay approximately 22.7 NIS for a one-unit increase in the use of public transportation.

The price variable in Model 2 is not significant because of a high correlation between the price and the constants.

​

Insights from the results:

Model 2 insights:

• Public Transportation: Increases probability of MaaS adoption. Aligns with Vij et al. (2020).
• Car Sharing & Minibus: Significant effect on utility after controlling for ASCs.
• Perceived Ease of Use: Significant for all packages.
• Privacy Concerns: Negatively significant, reducing likelihood of MaaS adoption.

​

• Custom vs. Pre-designed Packages:
• beta_Personal_Menu: Positive coefficient (0.292), higher likelihood for custom packages.
• Frequency_Job-Related Trips: Significant positive impact on adoption for all packages.
• Pack 1: 0.163
• Pack 2: 0.268
• Pack 3: 0.338

​

​

Insights from the results:

​

Demographic Insights

• Gender: Males less likely to choose Pack 1 and Pack 2 (Coefficient: -0.327).
• Young Age Group:
• Pack 1 & 2: Not significant
• Pack 3: Significant positive impact (Coefficient: 0.79, p-value: 0.0007)
• Old Age Group:
• Pack 1 & 2: Marginal negative impact (Coefficient: -0.349, p-value: 0.1)
• Pack 3: Negative impact (Coefficient: -0.46, p-value: 0.143)

​

Conclusion

• Significant Findings:
• Public transportation, car sharing, and minibus increase MaaS adoption.
• Privacy concerns act as barriers.
• Demographic Preferences:
• Young individuals prefer Pack 3.
• Older individuals are less likely to choose any packages.

​

​

Conclusions:

• Viability Testing:
• Initial findings tested the viability of our proposed methodology.
• Gathered initial "clues" on key factors affecting MaaS adoption.

​

• Need for Re-Planning:
• Deepen understanding of inclusive MaaS attributes.
• Action: Develop a latent variable of "MaaS inclusiveness.": Connect privacy/regulation concerns, data sharing, and accessible transport

​

Key Observations:

• Consistent trends observed with job frequency variable.
• Potential initial disinterest in integrated mobility.

​

Future Directions:

• Gain deeper understanding of elderly technical and physical abilities by Conduct focus groups/interviews with vulnerable groups.
• Explore TAM model elements: "usefulness" and "ease of use.“,

Investigate their correlation with MaaS inclusiveness.

• Explore preferences for "pay as you go" per ride.

.

Contribution:

1. User Insights and Inclusiveness:
1. Gain insights into user perspectives on MaaS adoption.
2. Provide recommendations to enhance inclusiveness in MaaS implementation.
3. Develop effective key recommendations for inclusive business models for MaaS deployment.

​

1. Methodological Advances:
1. Hybrid Choice Model (HCM) Framework

- Addresses inclusiveness for diverse user groups.

- Emphasizes legal, equity, and privacy considerations.

- Integrates TAM for a comprehensive understanding of travel behavior.

​

• Activity-Based Accessibility Model Extension

- Develop an Activity-Based Model of Inclusiveness (ABMi).

- Considers privacy concerns, technical abilities, and salary.

​

3. Expected outcomes:

• Travel Behavior Understanding: Advanced theoretical understanding of travel behavior.
• Policy Recommendations: Provide practical recommendations for policymakers and transportation planners.
• Inclusive MaaS Solutions: Ensure equitable access to transportation services by identifying specific requirements across demographics

Next step-

User Inclusiveness MaaS measure (UIMi)

Forgoing trips

Pay as you go

Private Car

Minibus

PT

MaaS – re-designed Package

PackB

PackA

• Initial implementation.
• We aim to add another dimension to the log sum by including inclusiveness indicators, Privacy concerns, technical abilities, salary, etc. Higher log-sum indicates greater inclusiveness.
• The logsum captures the expected utility of all available alternatives.
• “Intention to use” as a latent variable that affects the MaaS package choice.
• In the first stage: “Simple” logit model, second stage: Hybrid Choice Model.

Intention to use (Latent variable

Ease of Use

Usefulness

User Inclusiveness MaaS measure (UIMi)

Forgoing trips

Pay as you go

Private Car

Minibus

PT

MaaS – re-designed Package

PackB

PackA

• SP Experiment:
• Stage 1: compare all Alternatives.
• Stage2: In the case the respondents chooses pay as you go-> now he needs to choose if to adopt MaaS or stay without rides “forgoing trips”. to examine if respondents choose not to adopt MaaS due to personal difficulties..
• "Forgoing trips” refers to trips and activities that people want to undertake but can not due to various reasons, as identified by Luiu et al., 2017
• “Recognizing the personal and societal issues that travel problems may create, it is necessary to properly measure travel problems and identify who is most affected.” Singer & Martens (2023).

How MaaS can bridge these gaps in elderly communities, increasing inclusiveness by enabling desired trips (reducing the forgoing trips).

​

Intention to use (Latent variable

Ease of Use

Usefulness

Thank you for your listening

Participants:

A total of 300 respondents (59% males and 41% women) recruited via a panel company completed the questionnaire in Hebrew. 15% of the respondents were between the ages of 18-24, 20% between the ages of 25 and 34, 19% between the ages of 35 and 44, 18% between 45 and 54, and 13% between 55 and 64. There are 13% between 65 and 74 and 2% between 75 and 85% of the respondents have available private cars to use.

​