The well-known British-Dutch transnational company, Unilever, delivers consumer goods to 2.5 billion people in 190 countries around the globe. More than 150 thousands Unilever employees are working their best everyday to achieve their world changing purpose, to make sustainable living commonplace.

One of the focuses of Unilever in the path towards becoming more sustainable is to decrease packaging waste by making the packaging smarter and more efficient.


Unilever makes use of a machine learning based solution for modelling the packaging process. In order to allow for modelling of the new package types easily, there was a need to make the machine learning model more accurate and flexible.

In 2019, Machine2Learn was approached to perform this task. Machine2Learn first created a new scalable model which reproduced the results of the in-house model, and then accompanied it with a graphical interface to make it possible to interact with the model and then updated the algorithm to be able to learn from new examples.

By developing a model that can automatically be applied to datasets for other types of packing, the amount of manual labour needed to develop a new model is drastically reduced.

If you are interested to learn more about how to utilize machine learning to solve real world problems, we offer a brief workshop which combines a theoretical overview of machine learning with practical examples.

Find more about M2L Quick Start workshop