Colorado Solar Sales Booster Shane Rodgers Graduate Project 3.11.25

Overview This project uses Python and simple machine learning to identify high-potential ZIP codes for solar panel sales in Colorado. It demonstrates how data-driven targeting can improve marketing efficiency. Project Description The Colorado Solar Sales Booster project: • Analyzes solar irradiance, home ownership, and property values across multiple regions • Uses a decision tree model to predict solar potential • Creates visualizations to guide marketing strategy • Identifies specific ZIP codes and regions for targeted campaigns Running the Jupyter Notebook To run this project: 1 Install required packages: pip install pandas numpy matplotlib seaborn scikit-learn 2 Create a new Jupyter notebook or open the provided Colorado_Solar_Sales_Booster.ipynb 3 Run the cells sequentially to generate the analysis and visualizations Project Results Running this notebook will produce: 1 A data analysis of Colorado ZIP codes and their solar potential 2 Feature importance analysis showing which factors most affect solar potential 3 Decision tree visualization explaining the prediction process 4 Regional comparison across key metrics 5 Identification of top ZIP codes for targeted marketing 6 Saved PNG files of all visualizations for use in presentations Code Sample Here's a glimpse of the code used in this project:

Colorado Solar Sales Booster Project

This notebook demonstrates a simple machine learning approach to identify high-potential

ZIP codes for solar panel sales in Colorado.

Import required libraries

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from sklearn.tree import DecisionTreeRegressor from sklearn.model_selection import train_test_split from sklearn.metrics import mean_squared_error, r2_score from sklearn import tree

Set better visual style

sns.set_style("whitegrid") plt.rcParams['figure.figsize'] = (10, 6)

Create sample data (you can replace this with your actual data import)

def create_sample_data(n_samples=30): """Create a sample data file with Colorado ZIP codes for demonstration""" import random

# Set random seed for reproducibility
random.seed(42)
np.random.seed(42)

regions = ["Denver Metro", "Front Range", "Western Slope"]
data = []

# Generate sample data
for i in range(n_samples):
    region = regions[i % 3]
    
    if region == "Denver Metro":
        solar = round(random.uniform(5.3, 5.7), 1)
        owner = round(random.uniform(45, 68), 1)
        home_value = random.randint(400000, 900000)
    elif region == "Front Range":
        solar = round(random.uniform(5.5, 5.9), 1)
        owner = round(random.uniform(60, 80), 1)
        home_value = random.randint(350000, 750000)
    else:  # Western Slope
        solar = round(random.uniform(5.8, 6.2), 1)
        owner = round(random.uniform(65, 85), 1)
        home_value = random.randint(300000, 650000)
        
    zip_code = f"80{random.randint(100, 999)}"
    
    data.append({
        "Zip_Code": zip_code,
        "Region": region,
        "Solar_Irradiance": solar,
        "Owner_Occupied_%": f"{owner}%",
        "Median_Home_Value": f"${home_value:,}"
    })

# Create dataframe
df = pd.DataFrame(data)
return df

Notes for Presentation This project is designed for a PowerPoint presentation that focuses on: • How data-driven targeting can improve marketing efficiency • Which factors matter most for solar potential • Regional differences across Colorado • Specific ZIP codes to target for marketing campaigns The visualizations generated by this notebook should be incorporated into your presentation for maximum impact. Requirements • Python 3.7+ • pandas • numpy • matplotlib • seaborn • scikit-learn • Jupyter Notebook Author Shane Rodgers License This project is licensed under the MIT License - see the LICENSE file for details.