Spatial Data Lake

Spatial Big Data Dashboard
Project Overview
Network engineers need to look at a large amount(500 million data points per day) of call log data to understand the coverage status in order to perform network analysis and fix network issues. They need an internal visualization tool to replace their command-line query tools to enhance productivity.
My Contributions
In order to achieve problem/solution fit, I worked with clients to understand their problems and test potential solutions through rapid prototyping. At the end, a proof of concept visualization application was designed and implemented. The application demonstrated a much better user experience by leveraging the new mapping technology - vector tiles, which was adopted by the client to replace their older raster map.
After talking to client's team, we outlined the problem space as following:
‍End Users: Network Engineers.
‍User Pain Points: The legacy command-line tool takes a long time to generate reports and not flexible to enter different parameters. Needs to deal with different toolsets.
‍Paid Customer: Client's GIS Team Managers and Tech Solution Architect.
‍Customer's Goal: Increase Network Engineer Productivity.
‍Constraints: The tool needs to be accessed by 50 engineers simultaneously; It needs to be portable and cross-platform. Data is updated weekly nationwide.
‍Project Goal: Get customer buy-in.
DEFINE
User Research
By talking to some network engineers, we learned some pain points and how they finish their tasks with the current solution

Ideally, we need to remove the two breaks in the flow, and make the whole experience as an integrated one. I got together with our solution architect and engineers to focus on generating ideas to solve these two problems.
DESIGN
User Flow
user flow(before)
user flow(after)
In a brainstorming session, we agreed that there must be a map component to visualize spatial data,  charts to display statistical data and user input widgets to perform data querying to replace the old command-line tools.

I sketched different ideas based on our initial discussion. The basic idea is to use charts to visualize numbers like signal strength and use maps to visualize locations. Different components are put on a single page.

After two quick calls with clients, our design has evolved a lot based on their feedback. We learned that: the map is their major way to navigate, and pinpoint poor coverage area. Engineers have a set of queries that were be used 80% of time.

For the first feedback, We made a redesign, which leaning towards a map based dashboard, in which map becomes the major component to display and interpret the data. For the second feedback, we used several drop-down boxes pre-populated with most frequently used queries to save the engineers' time to manually type in queries with different parameters.
PROTOTYPE
Wireframing
Rapid Prototyping
Coded Prototyping
2nd iteration design
Coded prototype
The project had a tight deadline of just 2 weeks, but we managed to complete on time. Our clients were very impressed by our big data technology and the user experience it can enable.
BEYOND
Iteration
customizable dashboard
3D signal strength visualization
Conclude
The proof of concept was just the beginning. We continued to work with clients and created solutions to solve their problems. Technologies were the enablers, what kept clients engaged was the understanding of the underlining people problems. The value we created can be better perceived through the story of how end-users goals being achieved.

Want to get in touch?
Drop me a line!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.