Axis Of Insight

Inspiration

We were inspired by the scale and real-world impact of Kimberly-Clark’s challenge: selecting optimal mill locations using only public data. It felt like solving a multi-million-dollar logistics puzzle — with data as our only tool.

What it does

Our model identifies the Top 20 high-demand metro areas in the U.S. and recommends the best 3 mill locations that maximize demand coverage while minimizing logistics costs.

How we built it

1. Collected data from Census, ZIP DB, BLS, and ACS
2. Modeled demand per ZIP based on households, income, and size
3. Clustered ZIPs into segments (Premium, Mass Market, Low Income)
4. Focused on Mass Market ZIPs in top cities
5. Scored mill locations using a hybrid of cost, distance, and zoning factors

Challenges we ran into

1. Merging messy datasets from different sources
2. Defining what “optimal” means for site selection
3. Creating visuals that clearly communicate complex analysis
4. Balancing accuracy with performance in large-scale ZIP-level processing

Accomplishments that we're proud of

1. Built a complete pipeline from raw data to strategic recommendation
2. Created a clear visual framework that simplifies complex decisions
3. Recommended 3 sites covering 87% of high-demand ZIPs

What we learned

1. How to translate demographic data into real business strategy
2. Importance of clustering for market segmentation
3. How to evaluate trade-offs between logistics cost and demand reach
4. Data storytelling is just as important as the model itself

What's next for Axis Of Insight

We aim to enhance our model with real-time logistics data and broader infrastructure insights. Our goal is to develop a scalable site selection toolkit that businesses can use for strategic expansion, combining data science, geospatial analysis, and intuitive visualizations to drive smarter, location-based decisions.

Built With

• bls
• built-with:-python
• excel
• geopandas
• jupyter
• matplotlib
• numpy
• pandas
• powerpoint
• scikit-learn
• seaborn
• shapely