Optimize distribution and delivery with machine learning models for real-time supply chain planning
How a large, growing food retailer leveraged tailored machine learning models to improve the distribution of perishable micro packages.

INDUSTRY RETAIL

The client

The client is the nations #1 provider of farm-fresh fruit to businesses. This family owned and operated company launched in 2008 as a way to help companies provide healthier food options to employees, quickly becoming one of the fastest-growing companies in the U.S. year over year. Their business model is heavily dependent on technology to ensure they can efficiently deliver perishable produce at the highest quality, using local delivery carriers.

Overview

With rapid growth placing a burden on current processes, the client looked to Google Cloud and Pluto7 to help them scale their distribution operations. By harnessing Pluto7’s expert Machine Learning support, the retailer was able to successfully leverage Artificial intelligence models to determine the most optimized routing for delivery to their customers.

Challenges

Rapid growth created need to deliver more micro packages than ever before

Custom distribution process unable to scale quickly

Future growth and revenue limited by lack of scalability

Strategy

Utilize Google Cloud Platform to quickly and efficiently collect all related data

Use Machine Learning and Artificial Intelligence to generate highly accurate routing recommendations for micro package distribution

Optimize costs while maintaining customer satisfaction

Results

Accurate predictions of best distribution routing for each target market

A fully scalable solution for the entire distribution process

“If we can continue to do what we’ve been able to do until now, and scale with growth, then I don’t have to worry about keeping up with demand, and can think about creating new business models.”

-CEO, Disruptive Retailer

The customer has grown rapidly, delivering farm-fresh fruit across the country with customers ranging from stealth mode startups to large enterprises. It’s common to see their boxes in office cafeterias, meeting rooms and company micro kitchens.

What does it take to recommend a high accuracy micro package distribution routing?

Successful micro package distribution of perishable products depends on multiple factors: reaching the customer on time, avoiding congestion on routes and precise location delivery. Selecting the best route and the right local delivery carrier in the distribution network involves a lot of complexity, and rule-based logic must be updated continuously to reflect the constant changes in patterns. This is a classic problem that Machine Learning and Artificial Intelligence based solutions can solve. In order to come up with a highly accurate predictive model for routing recommendation, past data needs to be readily available and properly labelled. Once this has been achieved, the data can then feed into a classification model that predicts the best routing for a given set of input parameters- such as to and from location, number of packages, size of packages, delivery time, cost, distribution vendor, and so on.

“On the business side, we wanted to continue to explore Google Cloud to centralize our data to serve as a foundational platform for our ongoing ventures with our partner ecosystem of fruit farmers, distributors and carriers.”

- CEO, Disruptive Food Retailer

High accuracy routing achieved

With these objectives in mind, this disruptive food retailer partnered with Pluto7 to leverage Google Cloud Platform to transform their data. They harnessed analytics, Machine Learning and Artificial Intelligence to determine best routing for their perishable product delivery.

With Pluto7’s expertise, they were able to identify the right Google Cloud solutions for their needs and quickly developed a Proof of Concept that demonstrated routing with over 90% accuracy- meeting the needs of the route and carrier selection for each target market.

“Using Google Cloud Platform’s immense processing power with cloud storage, Cloud SQL, Dataprep, BigQuery and Data Studio, and leveraging its Machine Learning engine, meant that complex correlations between multiple sets of features could be processed quickly. This led to highly accurate distribution routing, ensuring farm-fresh products delivered on time to customers.”

-Salil Amonkar, Global Head of ML/AI Practice & Delivery

Stronger through innovation

Working with Pluto7, this client is experimenting with Google Cloud technology to leverage more data, analytics and Machine Learning. Ultimately, the objective is to achieve breakthrough innovation and business transformation by expanding the success of leveraging Google Cloud, Artificial intelligence and Machine Learning to related use cases and beyond.

Why Google

This client chose Google Cloud Platform because it:

  • Offers a flexible, scalable, ready-made infrastructure in the cloud
  • Provides a cost-effective platform that’s easy for marketing team members to use
  • Delivers speed, security, reliability and flexible pricing

Products used

Google Cloud Platform

Google BigQuery

Google Cloud Storage

Google Cloud Dataprep

Google Cloud Dataflow

Google Machine Learning

Tensorflow