High fidelity prototype for a mobile app for increasing pedestrian safety at night.


SafePath is a mobile application that enables the user in finding the safest path between two points on the map at night, specifically designed for U District area and University of Washington Seattle community.

Role: Product Manager, UX Researcher, Prototyper, Interaction Designer

Key Activities: Secondary research, primary research, ideation, interviewing, shadowing, cognitive walkthroughs, wireframing, prototyping, usability testing, interaction design, interface design

Initial Team Members: Abhigyan Kaustubh, Adrian D’sa, Carrie Wang, Gloria Deng
(Continued working on it independently on the side.)

Initial Timeline: Sept 2012 – Dec 2012, 10 weeks

Tools: JustProto, MS Power Point, MS Excel, Paper, Pencil


Executed UCD process to design the application and its experience.

Process Flow - SafePath (4)


Brainstorming and Planning

The initial steps involved understanding the scenario in which the user will be using this in, identifying the main needs of the user in that scenario, and transforming those into use case and features. We started this off via group discussion and lots of sketching (some of them are featured below).



We conducted interviewing and stakeholder analysis, the results for which are compiled here.

Screen Shot 2016-03-03 at 10.18.05 AM     Screen Shot 2016-03-03 at 10.16.14 AMA snippet of our gathered survey data.

Analysis and Requirements

Based on these, we got the following key insights in the form of user design requirements list:

1. The solution should show streets where criminal activities happen in real time, as people can avoid such places and choose another street.

2. The solution should be connected to the Police database containing historical and current records of criminal activities that have taken place in the University District, in order to retrieve this information for the benefit of the user.

3. The solution should update its data in real time in order to efficiently record crime and inform the user. 4. The solution should be connected to a city mapping interface, as it is easier for people to get the information of a street on a map than using a text description.

5. The solution should have a user tracking functionality, as people would require to fix their position on the map, incase if any incident occurs to him/her.

6. The solution should show security statistics of each street clearly, as people would want to easily judge which street to walk back home on.

7. The solution should tell user details of a street, such as the crowd density, luminosity and if there are any shelters nearby, as people want to know the elements that make them feel safe or unsafe.

8. The solution should have the ability to tell the users which path will be the safest from start point to their destinations according to the time of travel, as the user would want to choose another path back home if a crime has occurred on his current path.

9. The solution should be able to record crime in audio and video format, and synchronize this data with the Police database.

10. The solution should have an easy to use interface in order to cater to all types of users (novice to excellent) of technology and to an age group between 18 and 65.

Refined Design question Based on our research, we decided to refine our design question to get a sharper focus.

New design question: How can college students and young professors avoid dangerous street when walking home late at night in the University District?

Initial design question: How can you utilize the crowd to reduce criminal activities on the streets, when heading home late night?

After our research and discussions, we found out that the initial question was a bit ambiguous and did not reflect our theme clearly.

There was no mention of:

 The type of user group our product was focused towards (students, professionals, party goers etc).

 The locality where criminal activities take place.

 The medium used in returning home (walking, public transport, car, etc).

Hence, our design question was remolded according to the user research survey conducted and the personas defined for our product. It addressed the users who would be interacting with the product, and the location and the medium used to return home late at night.


Screen Shot 2016-03-03 at 9.23.36 AM Screen Shot 2016-03-03 at 9.23.48 AM Screen Shot 2016-03-03 at 9.24.03 AM

Key Features

Based on all the brainstorming and analysis, we chose 3 main features to incorporate in the application:

1. Gesture alert

Screen Shot 2016-03-03 at 10.31.00 AM

Alerting the Police via touch based gesture recognition software is an important add-on to our mobile application for smartphones. This feature aptly fits the requirement of our third Persona „Peter Professional‟. Using this feature, the victim can alert the Police in the vicinity through a signal sent by sliding his/her finger over the phone‟s display screen. The working model for this feature is shown in the sketch. Initially, the user has to set his/her unique gesture on the phone by clicking on “Add Gesture” button in the applications main interface. The application retrieves this gesture and stores it internally. The user will be allowed to change the gesture at any time while the application is running on his phone. Also, the user can customize a message such as „I am under attack. Please save me.‟ on the interface which will be stored in the application. Thus, when a user is under attack, he/she can signal the Police without getting noticed by the criminal. The Police will receive a signal containing the message saved by the user and his/her location information to help track the victim.

This feature will be beneficial for the user as he/she can signal the Police easily and secretly without getting noticed. Although the victim could face hostility in the interim, this application could promptly communicate to the Police while the crime is taking place. Also, since these gestures are customizable, a user can choose a gesture that is easy to remember and maneuver while under attack.

2. Data Collection

Data Collection

Data collection is one of the most important requirements of our design. Through this, we can collect the information regarding various criminal activities taking place in the locality. When users encounter any kind of misdemeanor, they can use this application to report their findings to others as well as to the Police. In this interface, we offer five features of describing criminal activities: site, type, extent of danger, description, and who the reporter wants to report to.

  1. Site Get the location of the user automatically or prompt the user to enter the location of the crime manually.
  2. Type Choosing the type of crime makes it easier for other users to figure out what criminal activity happened at that location and therefore, decide whether to choose an alternate route to avoid it.
  3. Extent of danger This shows the intensity of crime taking place in that locality. Users can choose a scale from 0-5 indicated by stars, where zero means low danger and five means maximum level of danger.
  4. Description Users can type in details about the crime. For example, how many of them were present, physical description, type of weapon used, etc.
  5. Report to Users can choose whom they want to report to. For a crime of relatively lower extent of danger, users can choose to report only to the application. For dangerous criminal activities such as mugging, stealing, murder, etc., they would also want to report the activity to the Police.
3. Path finding and Group Forming

The following designs basically encompass two ideas that have to be used in succession, the second one being optional and dependent upon the user‟s preference.

The first here deals with finding the path from source to the destination. The user can enter the source or it can be automatically be inputted by the system using GPS, which is a feature common in most cell phones. After this the user enters the destination where he needs to go, and then chooses the time he will be traveling at. Then the application will give the “live” path based on the parameters like well-lit paths, presence of people, residential areas, areas with 24/7 shops in immediate vicinity, roads part of or near main roads, etc. These parameters will be shown to the user who will have the freedom to choose (the shaded and encircled ones in the sketch) the parameters he wants to keep, and the paths will be shown accordingly. Also, this application will be connected to the Empathy Database and the Police Crime Record Database, and will be able to give updates about the paths in real time (recent criminal activities happened, criminal activity being reported by another user by Data Collection feature of the application), and then give choices and suggestions to the users accordingly.

This is the second part which is optional and can be used by the people who would want company of other people to walk at night. The user just has to click and hold the crowd symbol (the second one in parameters) and then can reach the above interface for forming a group. This grouping function helps to show people, who also want to be accompanied, on map as a red spot. If the red spot is clicked, it will show the information of those people (information that those people have pre-decided, just like people choose how much of their facebook profile is visible and to whom), so that the people can be identified. The information will mostly include the time when they will most probably travel on these street, their path and their message. This information can be collected by the person‟s facebook/ LinkedIn profile, and Contacts of the user himself. But this is the case when the user knows the other person. There can be situations in which there are other people who want to be accompanied and have no problems with walking with other people they don‟t know. To overcome this problem, the application will have its own private database wherein people who fall in this category can sign up or add their records, with a valid SSN/ Passport Number, so that the contacts authenticity is confirmed. Hence, clicking on the ‘Get contact’ icon, the user will be able to see the information that those people have made visible to others in the application‟s database. After this, if the user chooses, he can get in contact with them via message or call by clicking on “Send Request”, and then, depending upon mutual agreement and understanding between the two, groups can be formed.


After identifying personas, we performed rapid paper prototyping and sketching, and conducted user testing to come up with following hi-fi prototypes:

Screen Shot 2016-03-03 at 9.22.08 AM


Evaluation and Testing

For evaluation and testing, we created a hi-fi prototype using JustProto, which can be found at my GitHub page.

We performed user testing on our target users who fit our initial 3 personas, and improved the design though iterations.