Georgia Perakis – Professeur, Sloan School of Management, MIT, États-Unis
Nous vous remercions de confirmer votre présence.
Joint work with Lennart Baardman (ORC PhD student), Maxime Cohen, (recently graduated ORC PhD), Swati Gupta (ORC PhD student), Jeremy Kalas (EECS Undergraduate), Zachary Leung (recently graduated ORC PhD), Danny Segev (U. Haifa) as well as Kiran Panchamgam (Oracle RGBU) and Anthony Smith (formerly from Oracle RGBU)
Promotions are a key instrument for driving sales and profits. Important examples include promotions for grocery retailers among others. The Promotion Optimization Problem (POP) is a challenging problem as the retailer needs to decide which products to promote, what is the depth of price discounts, when to schedule the promotions and what vehicle to use to promote each product. This presentation will reflect our ongoing collaboration over the past few years with Oracle RGBU on promotion planning (from pricing to promotion vehicle selection).
An important consumer behavior we will incorporate and which is a direct consequence of promotions in grocery retail is that consumers stockpile the products on promotion and then experience promotion fatigue after the promotion ends. Unfortunately, the underlying optimization formulation even for a single product is NP-hard and highly nonlinear. We will first propose a linear approximation and show how to solve the problem efficiently as a linear programming (LP) problem. We will discuss analytical bounds on the accuracy of this LP approximation relative to exact problem solution. We will also consider a graphical representation of the problem which will allow us to employ a Dynamic Programming (DP) solution approach as an alternative. We will discuss the tradeoffs between the two approaches (LP vs DP).
Apart from the pricing aspect, we will consider how to decide which vehicle to use each time in order to promote which product. We will introduce greedy and integer optimization ideas in order to solve the vehicle selection problem in a tractable way. These methods are computationally efficient and hence easy to use in practice. We will also discuss some performance guarantees for these methods.
Together with our industry collaborators from Oracle Retail, we show that our models run fast in practice using actual data from grocery retailers and that the accuracy is high. We determine that they can improve profits by 3% just by optimizing the promotion schedule and up to 5% by slightly modifying some business requirements.
Campus de l'Université de Montréal