LUXURY

Estimating production costs for leather goods

#Web App

#Dataiku

#Pricing Model

#Ambition

Ambition
of the mission

In the world of leather goods, the stages of assembly research and profit study are numerous and slow before production is launched. Indeed, the choice of materials and components when creating a product involves a long phase of study by teams of experts. In order to optimize the time needed to calculate the cost of production and obtain a more accurate estimate, Eulidia has set up an algorithm for predicting the cost of production of leather goods and developed a web interface for teams.

#Method

Our
approach

Think

Speed of execution

The price of a leather goods product is analyzed in terms of “a single component” and estimating the price of each of these components and their assembly steps can be tedious. It was therefore essential to take into account this constraint and to rely on the benefits of AI to improve the process of putting products into production.

Make

Easy-to-use interface

Dataiku allowed us to combine a predictive model and a web application for this project. Through a web interface created on DSS, including the front-end and back-end, estimating production costs becomes a simple and fast tool for experts. The tool makes it possible to determine the price of each product at the component scale and allows the user to precisely measure the impact of the choice of incoming materials in the composition of the future product.

Scale

Better Management of Product Launches

The infrastructure and the model are ready to be deployed, with the ambition to thus become a key tool for future product developments. The launch of the production of products is more reliable and more in line with reality. This project is also intended to be used in other subsidiaries of the group, and to be extended to other product categories.

#Benefits

Indicators
of success

÷ 10 the error Between Estimated and Real Production Prices

Reduction Significant calculation time (4h -> 4min)

User interface Friendly

#DataStories

Discover our
other use cases

See all our use cases