Optimizing transportation of consumer packaged goods
Together with Unilever, we developed a machine learning solution that maximizes product packaging efficiency, enabling Unilever to get closer to zero defect transportation.
Unilever is a global corporation that produces a wide range of consumer goods. The company offers over 400 brand-name products to 3.4 billion people in 190 countries worldwide. Its team of 127 000 employees strives daily to achieve the company's purpose: to make sustainable living commonplace.
One of Unilever's goals in its journey towards sustainability is to reduce packaging waste by making packaging smarter and more efficient. To support Unilever in its quest for sustainability, we provided them with a web application solution to optimize how they stack the packages on top of each other so that the defect caused by transportation would be minimal.
We started by creating a new scalable machine learning model that accurately replicated the results of the existing in-house model and developed a graphical interface for the new model to facilitate easy interaction for Unilever personnel. In addition, we updated the algorithm to enable it to learn from new examples. The model can automatically be applied to datasets for other types of packing. As a result, the manual labor required to develop a new model is significantly reduced.
The web application:
- Helps to optimize the transportation of consumer packaged goods, providing Unilever with a cost-effective solution to optimize the packaging for shipments.
- Offers flexibility to planning engineers, allowing them to control various input parameters such as product type and maximum allowed weight through a user-friendly web interface.
- Minimizes transportation costs and waste by suggesting how to optimally pack products inside a storage container and subsequently stack the containers onto pallets.
,
Ready to level up your AI capabilities?
Succeeding in AI requires a commitment to long-term product development. Let’s start today.