Improving Manufacturing
Processes with Essilor

Today, predictive maintenance is widely considered to be the obvious next step for any business with high-capital assets, harnessing machine learning to control rising equipment maintenance costs. Predictive maintenance takes data from multiple and varied sources, combines it, and uses machine learning techniques to anticipate equipment failure before it happens.

Essilor is the world’s leading ophthalmic optics company which designs, manufactures, and markets a wide range of lenses to improve and protect eyesight. Essilor employs 67,000 people worldwide; it has 34 plants, 481 prescription laboratories and edging facilities, as well as four research and development centers around the world. In its dedication to ensure factories are efficient, innovative, and compliant with high quality standards, Essilor has a Global Engineering (GE) service that is responsible for the implementation and standardization of production processes.

Seeing that one of their goals is to find ways to better answer consumer and business needs, the GE team was facing the challenge of improving processes and performance of the surfacing machines to significantly improve their production by using the increasing volume of data.

We wanted a data science platform that would allow us to solve our business use cases very quickly. Thanks to Dataiku and its collaborative platform, which is agile and flexible, data science has become the norm and is now used more widely within our organization and around the world.”

Cédric Sileo Data Science Leader at Global Engineering, Essilor

Go further: Do More Machine Learning & AI (Without More Data Scientists)

Essilor chose Dataiku to help them effectively work with the extensive amount of data from the surfacing machines because:

• The setup and implementation of Dataiku was easy and allowed them to get started quickly.
• The team wanted one tool from start to finish that would allow them to explore, analyze, and create predictive models and that could be used by everyone, from professional experts to 
machine operators, data scientists and IT, and everyone in between.
• The data from the surfacing machines came in various formats, and some of the datasets were 
not reliable or were incomplete. Dataiku allowed them to manage these variations efficiently and 
effectively.
• Dataiku allowed to quickly test and iterate on use cases to arrive at a solution faster.
• The team wanted the flexibility to work using code or using the point-and-click visual interface, 
whichever sped up the work process.

How Essilor uses Dataiku

The GE team at Essilor was able to use Dataiku for a predictive maintenance use case. The goal was to be able to indicate to operators the right time to change the consumable components so as to optimize their lifespan, while also guaranteeing the production quality standards.

In addition to using Dataiku to predict the optimal conditions under which they will use the consumables completely, the GE team also leverages Dataiku’s web application feature to provide a visual decision tool to help operators make a consistent choice when they have to change the consumables. Additionally, the GE team provides value to Essilor as a whole by:

• Promoting the use of data to provide insight throughout the rest of the organization — Dataiku is now being used internationally by additional teams.
• Encouraging cross-team collaboration, as the use case was a collaborative work between the GE department, factories, and IT teams.
• Gaining a dedicated resource for data science, which will allow them to continue to make future improvements.

The next steps for Essilor will be to test their machine learning models in production, deploy dashboards across several labs, and roll out future improvements to surfacing using deep learning with images.