For my General Assembly (GA) Data Analytics Capstone, I analyzed the Maven Analytics shipping dataset to address a critical inefficiency at Candy Cache: 68% of orders shipping from suboptimal factories, wasting 7.5 million miles annually and costing $88K in lost profits and excess shipping.
Traditional rule-based analysis falls short in complex supply chains. I developed advanced machine learning on the shipping dataset, where my Random Forest model achieved 94.9% accuracy in predicting optimal factory assignments—transforming reactive logistics into proactive decision-making.
The solution includes an interactive decision-support tool that empowers logistics teams to simulate routing scenarios and quantify profit impacts before committing resources, shifting operations from guesswork to precision planning. This demonstrates how data science tranforms traditional supply chain management into competitive advantage.
Check out my Presentation and Tableau visualization for better appreciation:
- Maven Analytics: Supply Chain Shipping Analysis Project
Data Source:https://mavenanalytics.io/guided-projects/supply-chain-shipping-analysis