Android Wear App Prototype
UX/UI Designer, UX Researcher
User Research, Task Analysis, Ideation, Storyboard, UI design, development of wizard dashboard, Video Documentation
Sketch, Python, Adobe CC
reco is a wearable app prototype created for the android wear platform to help improve recycling quality on campus as well as to raise user's awareness regarding environmental sustainability. The app nudges users as they approach bin sets to segregate waste, then raises a few simple yes/no questions. By going through a few short questions, user will be able to segregate the piece of trash into the right bin hence gain knowledge regarding proper recycling practices.
This is a project accomplished in collaboration with my peers at Georgia Tech. In order solve the right problem concerning environmental sustainability on campus, our multidisciplinary team executed human centered design process well with detailed user research, ideation, interaction design and android app development.
i. Problem Scope
In order to understand what concerns the campus in terms of environmental sustainability, our group conducted preliminary semi-structured interviews with 10 novices (Georgia Tech's students) and 4 experts (Representatives of different GT organizations motivated to improve the environmental sustainability on campus).
Both types of preliminary interviews pointed us into the direction of recycling. We also received waste audit data from waste management office identifying 41% of waste were improperly disposed in the campus multi-stream recycling system. Therefore, we decided to move on to focusing waste segregation on campus as our problem space, and define the target users as people on campus who are concerned towards segregating waste correctly but do not have the required knowledge to do it properly.
ii. User Research
In order to understand the strategies and challenges behind the waste segregation by students on campus, we conducted the following the research methods:
We observed 25 students across various locations around campus as they threw away their trash into several waste bins around campus to learn how students actually dispose their trash.
Task based Controlled Observations
5 students were recruited to complete a waste sorting task on campus. We implemented thinking-aloud protocol in the process. Participants had to justify their rationale behind the waste segregation decision and report how confident they are of their action after sorting each item.
The result of our observations helped us to empathize with our target users better. The observations helped me to structure a solid task analysis diagram, so that we can identify the key-points where we can introduce a design intervention to address the problem. By analyzing the results, we came to reach the following user needs:
With the user needs identified, I developed the two personas to be referenced and empathized with in the design phase.
i. Diverging phase
Our research helped us to identify how users’ waste segregation practices are significantly influenced by two factors: knowledge and convenience. With these understandings, we focused on introducing design solution that caters those needs. We had a brainstorming session for each group member to generate multiple ideas, and by the end of the brainstorming session we ended up with 21 unique ideas.
ii. Converge to top 4 Ideas
In order to sort our ideas effectively, we introduced a classification system in a chart form with x & y axises which were mapped according to human behavior in chronological phases and a user-system scale. At the end we reached 4 distinct solutions which tackle the waste segregation from a different angle. Low-fidelity prototypes of these 4 ideas were made to gain constructive feedbacks.
iii. Prototypes of top 4 ideas
1. Lock screen training
A lock screen utility which requires users to sort virtual waste to one of the bin categories to unlock the screen. An accuracy report will be generated to inform users regarding their progress/improvement
2. Wearable Nudge
A just-in-time wearable app responding to the bin the user is closest to, and nudges the user to think about his waste sorting decision.
3. AR educational game
Mass AR game requiring users to capture as much recycling centers as possible on campus by sorting waste correctly.
4. “I don’t know” bin
Novice user with difficulty sorting waste can leave their trash in an “I don’t know” bin. An expert user will help to sort the waste by setting the mechanical claw installed on the IDK to dispose the item to the correct bin remotely.
In order to select the best design to resolve the problem of waste segregation for our target users, we defined our selection criteria in four main areas listed as following:
Just In Time: A design intervention will be the most effective when the action takes place right in time. This guides the user to make the right decision at the moment and requires low cognitive load from the user
Convenience: As our user research indicated convenience as a key factor influencing users’ decision, we decided a solution which does not make users feel inconvenient in his daily routine is essential. The design should also be easy and quick to use
Knowledge: We want the users to do the right thing and also gain knowledge from the process, so that they can learn to do the right thing on their own in the future
Behavior Change: Our goal is to introduce a truly sustainable solution which evokes a change in user behavior. Therefore our design has to help the user to make the choice and guide them to the right path by building knowledge
Based on these selection criteria, we decided to select Wearable Nudge Smartwatch App as our final idea. We reached this decision as it is a just-in-time, sustainable design solution welcomed by multiple stakeholders which educates users through the process. As we have decided, we renamed the project idea as reco.
i. How does it work
The following storyboard illustrates how does reco work
We kept the following design considerations in mind when developing reco:
1 Aesthetic & Minimalist UI - Since our aim was to help the users complete task efficiently, we kept a simple aesthetic with basic interactions to maximize convenience for our users
2 Consistent & Concise content - We ensured all questions are structured similar form (short and to the point) to minimize cognitive load processing the information.
3 Visibility of system status - The state of flow needs to be effectively communicated to users using visual icons and color
iii. User flow
With the design considerations in mind, we moved on to developing the user flow with a decision tree shown as following:
iv. Final Build
Two areas of development are the wearable app itself and proximity sensors to be installed on the bins to trigger reaction in the nudging app. Initially we planned to develop the wearable app with Java and the proximity sensors with Arduino. However, due to the limited time frame given, we decided that sensing the proximity between user and bin and then triggering a signal to the user's device can be replaced by a Wizard-of-oz technique as it would have achieved the same result in the usability testing phase.
Therefore, after I finished designing the UI, I moved on to develop an auxiliary wizard dashboard with python for our researcher to send signals and trigger a reaction on the wearable. I also took charge of designing the presentation poster as well as filming the video documentation to tell the story of our users as well as the story about how this project is developed.
At the end we delivered the following for our project:
4. Research Plan
i. Proposed study
As our team is still waiting for an IRB approval, we have not been able to conduct any usability testing yet. However we have already constructed a proposed study involving 40 participants in a between-subjects study (divided into experimental group and controlled group. The experimental group will complete task with reco while the controlled group will not) to measure our success metrics in the following hypotheses:
H1 (comparing accuracy) - The overall accuracy of sorting waste will be significantly higher for the group using the reco compared to the controlled group.
H2 (comparing convenience using task time) - The overall task time will not be significantly different between the two conditions.
H3(comparing knowledge) - Experimental group participants should be able to sort the hypothetical waste more accurately in the post test after using reco.
Figuring out the right problem
We had a hard time locking down our problem because we want to find out a real problem in the area instead of just jumping to a problem we assume our users have. We did a lot more preliminary research and spent much longer time to narrow down our problem scope than other peers. Yet all of our team members felt extremely rewarding when we narrowed down our problem scope to waste segregation, and became more determined to design a practical design solution - because research told us it’s a real problem concerning our stakeholders and it is a problem worth solving.
Disputes are good
Our team members did not see eye to eye on everything at first. For instance we spent a long time arguing when should the design solution be introduced - should it be the time when users want to dispose a piece of trash and cannot find a recycling bin? Or should it be the moment the user actually disposes a trash in a multi-stream system. Yet the disputes led us to ask the right questions in research phase, and gave us a better insight to work on our project. These insights then united us a team in an even better way as we overcame the obstacles together.
Oz it when you need it
A lot of times designers find themselves struggling with a limited project time frame. We encountered the same issue and at that time we decided evaluating usability is more important than making the system fully functional. Therefore we changed the triggering factor of reco from sensor activation to wizard-of-oz at the development phase, so that we can move on to the phase of evaluation faster.
Other Selected Work
UX Design & Research
Service Design & Research
UX Design & Research
UX Design & Research
UX Design & Research
THEATRE OF MADNESS
Mixed Reality Design