How a Real-Time Delivery App Was Built in a Month

delivery optimization

How does a company make certain packages arrive on time?

Name of Organization: Chronopost

Industry: Postal

Location: Antony, France

Business Opportunity or Challenge Encountered:

With the advent of e-commerce retail, delivery companies have seen their activity increase significantly in recent years. Many shipments may take up to a week or more, but some companies tout two-day delivery as a value-added option.

Chronopost International, a member of the La Poste Group (French Postal Service), is no exception. It provides express shipping and delivery services both domestically (in France) and internationally. In 2013, Chronopost transported 102.2 million packages in over 230 countries in Europe and worldwide. Chronopost maintains a promise: that all parcel deliveries in France will arrive by 1 p.m. the day following an order. But as demand continues to grow, and especially during critical periods such as Christmas or Mother’s Day, the company had trouble keeping that guarantee. The company needed to be able to apply historical data to real-time service to better optimize delivery operations and ensure delivery deadlines.

 How This Business Opportunity or Challenge Was Met:

To meet the need for real-time awareness of delivery challenges, Chronopost created and implemented an application that automatically generates an ease-of-delivery rating for each address. The application, built with Dataiku Data Science Studio (DSS), takes into account historical internal delivery and retrieval data. It analyzes and enriches shipping and delivery data via data aggregation by geographic location. Eventually, it enables easy modeling to score each delivery by time, location, and date — therefore optimizing the delivery route and reducing costs.

The incorporation of new deliveries to the existing model enables iteration and continuous production cost optimization. To implement this project on Dataiku DSS, Chronopost leveraged technologies including SQL, JavaScript, Python, and used both in-house and specific scoring models. It took one month, one data scientist, and a CTO to deploy the first version of the application.

The delivery optimization application enables the company’s global network of business intelligence teams to collect and use performance indicators on demand. The operational means and costs involved in package delivery were optimized and Chronopost even managed to create new commercial offers based on delivery timeframes.

Measurable/Quantifiable and “Soft” Benefits From This Initiative:

As a result of the new system, Chronopost claims it can ensure constant quality of their different offers (delivery before 1 p.m. or before 8 a.m.) at optimized production costs.

Using and creating this new application has opened the door to new opportunities for Chronopost’s commercial, organizational, and operational BI strategy, says Régine Buys, Chronopost’s business intelligence manager. Being enable to easily access, analyze, and use data in a framework combining both standard and Big Data architectures has enabled Chronopost to improve knowledge management, as well as customer satisfaction.

(Source: Dataiku)


A Waze for fleet fuel management

Archives: customer experience management

Leave a Reply

Your email address will not be published. Required fields are marked *