Distribution Network Visualization

• 

  Baseline Scenario

• 

  New Scenario

• 

  Impact Assessment

Inspiration

Opportunity to work on actual industry problem. Additionally, showcasing visualization and problem solving skills.

What it does

Visualizes current and new proposed network.

How we built it

I mainly used Python (with Github copilot) to find solution to the problem and Tableau to visualize it. Gurobi software's facility location problem tutorial helped me learn how to find the next best location.

Challenges we ran into

Input data needed some cleaning as it had some invalid values (e.g. TO_ZIP had text value as "zip code"). Calculating actual driving distance from New candidate locations to customer locations was a challenge as I did not find an easy tool/service that could be used. Ultimately, I settled on using Geodesic distance and applied a factor as estimated using linear regression to estimate actual driving distance.

Accomplishments that we're proud of

Learned new algorithms and use Python libraries like Gurobi. I was able apply linear regression technique on input data to estimate a factor to calculate travel time from distance for new candidate locations to customer locations.

What we learned

Lot of new capabilites of Python and Tableau. GitHub copilot made it very easy to code in Python. Visualizing the data pretty much gave away/narrowed down the area for next best location.

What's next for Distribution Network Visualization

Excited for future hackathons.

Built With

• python
• tableau