Canada Trade Promo Optimization (TPO) Transformation: Modernizing the End-to-End Pipeline

(1)
The main activities of the partner
Unilever Canada is a global leader in the consumer packaged goods (CPG) industry, focusing on innovation and sustainability across its five business groups: Beauty & Wellbeing, Personal Care, Home Care, Nutrition, and Ice Cream. The company integrates AI and data-driven approaches to optimize supply chain processes, improve consumer insights, and enhance business operations. A key focus is Trade Promotion Optimization (TPO), where Unilever applies machine learning techniques to enhance promotional efficiency. In this project, Unilever will provide industry expertise, access to proprietary datasets, and business insights to ensure the research aligns with real-world applications.
(2)
The challenges the partner aims to solve through this project
Unilever faces challenges in scaling AI-driven trade promotion strategies across its diverse portfolio of 400+ products. Traditional trade promotion planning is inefficient, requiring extensive manual effort and struggling to balance multiple business goals, such as maximizing ROI, increasing market share, and reducing trade spend. Existing machine learning models also suffer from computational inefficiencies, limiting scalability. Additionally, integrating multi-objective optimization algorithms with real-time data analytics remains a complex challenge. This project aims to refine Unilever’s predictive models and streamline the end-to-end ML pipeline to enhance decision-making efficiency.
(3)
The anticipated social or economic benefits of the project for the partner organization(s)
For Unilever, this research will lead to the development of an AI-powered TPO system, enhancing decision-making, reducing operational inefficiencies, and improving promotional effectiveness. Automating trade promotion planning will result in cost savings, higher profitability, and better allocation of marketing budgets. More broadly, society benefits from optimized trade promotions that improve product availability, create cost savings for consumers, and contribute to a more sustainable and efficient retail supply chain. Furthermore, advancements in AI-driven decision-making from this project could set a benchmark for other industries seeking to improve operational efficiency through machine learning.

Faculty Supervisor:

Eldan Cohen

Student:

Partner:

Unilever Canada Inc

Discipline:

Computer science

Sector:

Manufacturing; Wholesale trade

University:

University of Toronto

Program: