International Choice Modelling Conference, International Choice Modelling Conference 2017

Font Size: 
Stated Preference Design for Exploring Demand for “Mobility as a Service” Plans
Melinda Matyas, Maria Kamargianni

Last modified: 28 March 2017

Abstract


The transport sector has recently been swept with a wave of innovative mobility solutions to increase its efficiency and eradicate dependence on private vehicles. One of these ideas, is the Mobility as a Service (MaaS) concept. This user-centric, digital and intelligent mobility distribution model aims to meet users’ transport needs via a single platform and offer all the services through a single mobility operator. MaaS aims to integrate all transportation modes (public transport, bike and car sharing, taxi etc.) and provide them to users through a single interface allowing purchase either as a pay-as-you-go service or as tailored monthly mobility plans including fixed amounts of each transport mode. Packaging, or bundling, services in such a way may be new to the urban transport sector, but has been commonly used in other sectors, such as telecommunications. As a new concept, there is still a gap in understanding how to create these mobility plans so that they can cater for the preferences of all the socio-demographic user groups. Against this background, the purpose of this paper is to first, describe the design of a stated preference (SP) experiment that captures the complex decision making process of purchasing MaaS products and, second, to validate this design using focus groups.

The design uses a prompted recall smartphone based travel survey tool (FMS, Cottrill et al., 2013), which is expanded by a SP experiment regarding MaaS monthly mobility plans choice. After filling out a pre-survey about basic sociodemographic information and mobility tool choices, respondents are tracked for a 7-day period. During the span of the tracking, they are reminded to verify their travel and non-travel activities and answer additional questions about their experiences (completed either on the web interface or their smartphones). As the case study area is Greater London, all the elements of the survey are adapted to fit the local environment. After the 7 days of tracking is complete, the revealed preference (RP) data is aggregated and users are presented with a summary record of their mobility behaviour for that week, broken down by transport mode and including information about travel-cost, time, and distance and number of trips. This, so-called Mobility Record (MR) is presented to users alongside the a description of what MaaS is to lead up to the SP. Further, a shortened version of the MR is shown alongside the SP, to allow users to reflect on their current travel habits and then make a choice in the experiment.

The SP has 4 alternatives: three fixed plan alternatives and one menu option where the users can determine which and how much of each mode they want. These are presented alongside each other, but only one of them can be chosen. Thus, the outcome of a choice made from the options is either one of the three plans or any combination of the individual features in the menu option. The flexibility of the menu option is priced, meaning that this plan is always more expensive (relatively) to the fixed plans. This approach is chosen to allow for analysis of peoples’ willingness to pay for flexibility within MaaS plans. The core attributes in the plans are the transport modes: public transport (with two levels unlimited bus, unlimited PT to match with the existing monthly pass options offered in London), bike sharing (yes, no), car sharing (with levels denominated in time), and taxi (with levels denominated in distance). As we live in the era of mass customisation, individuals are used to services that are personalised to their needs. Taking this into account, the SP was tailored to each respondent to present plans that suit their specific requirements. This contextual information is gathered from the RP sections of the survey, the pre-questionnaire and the activity diary, as well as the mobility record. The plans are tailored in two ways. First, the decision to include or exclude certain mode-attributes are based on prior knowledge of the user. For example, if they stated that they do not have a licence, they the car sharing attribute was excluded from all their plans. Second, the attribute levels of some mode-attributes are pivoted off their mobility record values (except if this value is zero in the MR). This is especially relevant for the taxi and car sharing attributes, which without this, could assume values that are completely unrelatable for the respondent.

Besides the core, mode based attributes, additional ones were also included in the SP to test for respondents’ openness to these features. These include ideas like ’10-minute taxi guarantee’ and ‘adding additional drivers to the car sharing plan for free’. Further, an incentive attribute was included with levels such as free food or grocery delivery for the month. The final two attributes are transferability, meaning how much of left over mode-attributes can be transferred to the next month, and the price attribute. The levels for the latter are fluctuated around the sum of the existing base prices of each attribute, which, where possible, are taken from their real life price or average price.

Even though the design is complex, the way it is presented to users is clear and as concise. Various design and presentation options were tested in three waves of focus groups, including individuals from a range of sociodemographic backgrounds. The first and the third focus groups were smaller (5 individuals) while the middle one was with around 20 individuals. These took the format of both email feedback and personal interviews (in some cases the combination of both) about preferred design, presented information and even wording.

This research aims to contribute the first step in the process of understanding peoples’ preferences for MaaS plans. The presented design combines several methods, e.g. smartphone activity diaries, pivoting, menu and simple choice into one experiment allowing us to efficiently examine the complex decisions potential future MaaS users will have to make. Even though the design was applied to London, it could be used to study MaaS in any other environment as well.


Conference registration is required in order to view papers.