An interactive kiosk to take the shuttle system to the next level

The CMU shuttle system aims to be a safe, reliable transportation option for students and faculty members to get to and from campus. CMU GSA approached our team in hopes of improving the shuttle system since various issues such as inability to gather comprehensive shuttle information and difficulty in finding the physical shuttle stops are still very prevalent in the community. From several rounds of research, our team recognized that a major trend across users is the lack of easy access to information about the shuttle service.


The MOAI is a physical kiosk located at CMU shuttle stops that riders can easily locate and interact with to discover real-time shuttle information like current shuttle capacities. On top of being a central resource for shuttle information, it also has an integrated check-in system that maps out the shuttle journey for drivers and ensures that riders have a stress-free journey to their destination.

Our Approach

  • Background Research
  • Think-Alouds
  • Usability Aspect Report
  • Walk the Wall
  • Reframing Activity
  • Focus Definition
  • Stakeholder Mapping
  • Generate

  • Research Plan
  • Contextual Inquiry
  • Affinity Diagramming
  • Surveys
  • Speed-Dating
  • Low-fi prototype
  • Evaluate

  • 5-second Testing
  • Experience Prototyping
  • Poster Presentation
  • Research Report
  • My Role

    As the Head of Research, I oversaw all research operations, defined research goals for the team, and approved research details and participants. This was carried out through regular progress check-ins with team members, accommodations for unexpected situations, and reminder messages to the team for deadlines. I also documented research tasks for each team member and made sure that each team member clearly understood their research purpose. Finally for participant recruitment, I ensured that the participants would contribute information pertinent to our project focus and that they were willing to cooperate in our research.

    We were mostly able to divide the work for each research method equally, but still had to split up some of the work. Notably, I spearheaded the affinity diagramming with Roxanne and took charge of logistical coordination for experience prototyping with our physical kiosk.

    • Project Duration5 weeks
    • TeamConnie Chau, Emily Ding, Tiffany Liu, Antonio Song, & Roxanne Zhang
    • Skills/ToolsStakeholder Analysis, Contextual Inquiry, Affinity Diagramming, Speed Dating, 5-second Testing, Prototyping

    Main Screen

    Glanceable bus schedule, arrival time, capacity, and an easy check-in scanner

    Route Details

    Real-time map with GPS locations and route overviews

    Route Selection

    Easy check-in with details about shuttle arrival times and capacities for accurate route planning

    Stop Selection

    No need to notify the driver at your stop anymore. Pre-select your destination and ride with peace of mind.


    "I find the information on the RideSystems app often inaccurate."

    "I decide whether to take the shuttle based on how long I will have to wait."

    "I have to constantly check my location on Google Maps in order to get off at the correct stop."

    Have you ever had such a bad first experience that you never wanted to try something again? This is the exact feedback that we received from many amateur CMU shuttle riders, who were scared away by multiple parts of the shuttle riding experience—from inability to find shuttle stops to conflicts with the bus driver because riders did not know when their stop was. Our team set out to use various research methods to understand the root cause of these pain points and innovate on an empowering solution.


    Background Research
    "I don't know what I don't know"

    Initial research from both quantitative sources such as the RideSystems app data and qualitative data gathered from contextual inquiries and think-alouds revealed five overarching key insights:

    1. Decrease in overall ridership across each fiscal year, but notable differences in the ridership of various shuttle routes which could reflect the changes in needs of those who live in certain areas around campus

    2. Heavy shuttle ridership during 8 to 10am and 6 to 9pm, reinforcing the notion that the CMU Shuttle Service is essential for students’ transportation needs to and from campus

    3. Riders are unsure where stops are located even when displayed on maps on mobile transportation apps (Tiramisu and RideSystems) and on the CMU shuttle website

    4. Riders’ transportation needs depend on three important circumstances: convenience, weather, and perception of accurate, updated data

    5. Students consistently report referencing another student’s involvement in their own decision to take the shuttle to not only learn about the shuttle service overall, but also the complementary apps and other tips and tricks about the service

    Based on these insights, our team determined that a major pain point for our student riders was a lack of information about the shuttle system overall, the state of the shuttles they wait for, and even where shuttle stops exist.

    Walk the Wall

    Since most of the background research was done individually, our team decided to look at all of our research reports and affinity diagrams together to review, share, and discuss the findings so far. Through this activity, we came to a common understanding of the CMU shuttle system with these insights:

  • Problems with the CMU shuttle services fall into two categories: physical experience and mobile application navigation.
  • Users experience a high level of uncertainty during their shuttle rides.
  • Existing apps (Ride Systems and Tiramisu) are difficult to use.
  • Riders do not consider the shuttle a reliable form of transportation.
  • Riders learn about shuttle information based on unofficial channels.
  • Riders decide on transportation options based on their priorities.
  • Riders decide on transportation options based on their priorities.
  • Reframing Activity
    Reversing our Assumptions

    In order to generate new, innovative ideas within the space of student transportation, we identified our pre-existing assumptions about the CMU shuttle system and generated the statements in reverse.

    Focus Definition

    In order to generate new, innovative ideas within the space of student transportation, we identified our pre-existing assumptions about the CMU shuttle system and generated the statements in reverse.

    Understanding our Stakeholders

    We sat down together to brainstorm stakeholders and think about how they relate together in the bigger system. This helped us identify potential users of our proposed solution, and how a solution may have bigger implications when considering the relationships between stakeholders.


    Our Research Plan

    Our team defined our research goals, questions/hypotheses, and methods table to help us consider what we want to proceed with in our research plan. Then, we selected aligning questions and methods based on their importance and benefit respectively. The topics we wanted to focus on were crowdsourcing methods and real-time data collection, and we wanted to find out how users may benefit from information from various inputs as well as how they might use information in different scenarios.

    Contextual Inquiry

    To understand the shuttle riding experience of our target groups, we followed both frequent and infrequent shuttle riders on their complete shuttle riding journey to take notes and observe.

    We sorted the interpretation notes from the contextual inquiry through affinity diagramming to identify general categories of needs and key user pain points. These insights became the starting point for our proposed solution.

  • Riders turn to peers and digital apps to find solutions to their shuttle questions because those platforms are easiest to access as students
  • Riders have to refer to different platforms for different aspects of shuttle information, and no single app/website could fulfill all of their needs
  • Riders prioritize real-time information updates about different aspects of the shuttle so that they would be less likely to experience unexpected breakdowns while on or before getting on their shuttle ride
  • Riders can make better decisions if provided with shuttle information before and during their rides
  • Different students have different degrees of comfort with a shared information network about the shuttle system, but they are all willing to be a part of it somehow by either contributing or gaining data out of it
  • Survey

    In order to gauge interest in and test feasibility of our proposed crowd-sourcing solution, we conducted online surveys to identify rider interest and willingness to be a part of a crowdsourced shuttle riding network.

    Based on survey results, we derived the following insights:

  • Most survey takers are receptive to using a crowdsourced solution (90%)
  • Not all information benefits from crowdsourced data
  • Interest in such a solution is high, but willingness to contribute is at different levels depending on the scope of crowdsourcing
  • Willingness to contribute depends on effort required and whether or not the input is objective or subjective
  • From this point on, we decided to pursue a limited crowdsourcing solution that does not require much input effort from riders but still can provide provide information on the popular areas of interest (shuttle capacity and shuttle delay).

    Speed Dating

    Each team member presented multiple storyboards to users in order to trigger discussion about the proposed scenarios, which helped us gather fast feedback and generate new ideas.

    We found that riders prefer cheap, simple solutions that are free and less time consuming. Users also value real-time and accurate shuttle information but are not majorly concerned about privacy.

    Low-fi Prototype

    As a team, we selected and voted on the top concept we wanted to move forward with: a digital touchscreen that will be a part of a shuttle bus stop. With this approach, users are able to have access to the information they need in a simple and straightforward manner to prepare for their upcoming shuttle journey.


    5-second Testing

    We tested our low-fidelity prototypes with our target users to quickly assess ease of use and identify areas of improvement. Through this method, we confirmed that people perceive the app interface to be about transportation and, as we had hoped, specifically a transportation app about the CMU shuttle service. However, user interest in interacting further with our screen was average—some participants seemed open to it but only a few seemed willing to do so on their own.

    Fast Feedback for Fast Improvements

    Our team iterated on the low-fidelity prototype to make refinements based on the 5-second testing feedback:

    Experience Prototyping

    Previously, our 5-second tests focused only on the interface component of our proposed solution. With experience prototyping, we wanted to test the check-in system and physical aspect of our proposed solution in the form of an object that users can interact with to understand how it may be integrated in a space.

    From multiple rounds of testing, we gathered the following insights about user needs and our current prototype:

  • The main screen is clear on its own because users can find necessary information for their trip, but it did not provide clear indication on where/how to check-in (which was one of our lower-level goals for the prototype)
  • Users were unclear about what the check-in process actually does
  • Users want a quick way to check-in, and the idea of potential long lines to check-in every time might cause frustration and concern
  • Final Prototype

    Our final solution culminates as an interactive kiosk that targets all of our main pain points found through multiple rounds of research:

    1. The interactive interface condenses all important information in one place, with a focus on real-time, accurate data

    2. The physicality of the kiosk increases awareness of shuttle stops and the CMU shuttle service as a whole

    3. The integrated check-in system allows both riders and drivers to determine stops ahead of time, eliminating the need to struggle with broken bus pull-cords or yell at drivers

    The final feature also provides passive crowdsourced data such as shuttle capacity for important stakeholders like drivers and the CMU transportation system. All in all, our solution addresses our focus while also affording opportunities for future development.

    Digital kiosk main screen with real-time updates on shuttle route times and capacities (left)

    Information page with instructions on how to ride the shuttle (right)

    Map view of selected shuttle route

    Select a shuttle route after scanning your ID card

    Scroll through shuttle stops to select your destination

    Confirm your stop to complete the check-in process

  • The more research methods, the merrier! I usually have the tendency to use only a few research methods per project because I want to get feedback only when I feel like I'm ready for it, but more feedback never hurts the process. In fact, the interim feedback I received on partially thought-out ideas helped push those ideas to success.
  • Though it may feel tedious in the moment, writing out clear research goals and procedures for each method conducted is crucial to getting the team on the same page and for reflecting later on. I wrote pages and pages of details for this project which felt quite unnecessary at the time, but was a blessing when I needed to explain all of our design decisions at the end.
  • More Projects