McDonald’s, one of the world’s most ubiquitous fast food chains, wants to “democratize” machine learning and data science practices across the company.
The fast food giant wants to enable end-to-end automation and machine learning from control to cost control.
The Fast food chain took a big leap toward ML when it launched DataBricks early last year for its data science and machine learning users, learning best practices on how to bring data to the platform.
It puts all the data into an s3 bucket with a data lake enabled, helping with data versioning and also building scalable and powerful engineering pipelines in the platform.
In less than nine months since the platform was launched, the company has grown from zero to the production scale of ML operations.
McDonald’s data science and analytics division was able to deliver more than 15 use cases and put more than 30 models into production before deploying them in the five countries in which it operates.
Several ML deployment frameworks have also been developed for data science and end-users to ensure that the fast food chain has adequate security frameworks for building and deploying ML models.
And while McDonald’s plans to expand the use of ML this year, the fast food giant expects to see at least four to five times that number.
Future tasks for ML include performing fine-grain projections, SKU-level forecasting for its restaurants, automating marketing and personalization activities beyond “good machine learning for marketing,” such as unmanned end-to-end interaction with customers through offerings, and cross-channel performance measurement across all of its markets.
For more information, read the original story in ZDNet.