The Story

Our client sought to optimize delivery routes for vehicles with limited capacity to improve the timely and efficient delivery of customized print products. They required a scalable ML model for optimal route selection, addressing challenges around estimating package weights, demand, and vehicle requirements while minimizing delivery costs.

Our client is a Swedish unicorn software company operating in 34 countries. They provide on-demand and optimal print services to global customers like Canva, Shopify, and Etsy, ensuring fast and efficient delivery.

Amit Jangid

Founder & CTO

Challenges

• > Determining the best delivery routes for vehicles with limited capacity.

• > Estimating package weights, customer demand, and moving distances.

• > Ensuring vehicle capacity was not exceeded while minimizing costs.

• > Finding a solution that could scale efficiently.

The Solution

Our team built an ML model to address the Capacitated Vehicle Routing Problem using Python, Solver Pulp, and Graph theory. Data analysis estimated package weights, demand, and vehicle requirements, while Google Maps API and Matplotlib were used for location visualization. We ensured data security and scalability through the S3 bucket framework. The result was a comprehensive solution that optimized route selection and improved overall delivery efficiency.

The Outcome

The ML model achieved a 60% cost reduction by optimizing vehicle requirements and moving distances, significantly enhancing delivery efficiency and cost-effectiveness. This solution enabled the client to meet operational excellence, ensuring timely deliveries and high-quality products to customers worldwide.