2020-12-28 | Sheng Bi: Promotion Design with Path Dependent Network Effects
2020-12-28
Abstract
The limited-time “collect and win” games are often used to promote products and drive sales to theretail outlets by creating short term temporal changes to customers' purchasingbehavior -- the desire to buy products on promotion and frequency of purchasesboth increase with the number of previous purchases. We study the promotiondesign problem for these games, to determine the set of eligible products andthe duration of the promotion. The customers' purchasing behavior depends notonly on the product attributes and features (static effect), but also onproduct eligibility for promotion and historical purchases (path dependentnetwork effect). We model the dynamic choice processes using poissonization ofthe Polya Urn models, to capture the transient change in the frequency ofpurchases and purchase probability of each product on promotion. We use thisapproach to study the optimal promotion design problem under different “collect and win” game settings, by solvingnon-convex assortment optimization problems. We obtain an exact and/orapproximation approach for these problems, and show that the revenue-orderedpromotion set is already near-optimal in many of these games. The optimalduration depends on the promotion set chosen, and also on the targeted numberof products sold before the game found a winner for the grand prize. Using aset of data provided by a fast-food company, we show the importance ofcarefully choosing the promotion set and promotion duration, both decisionsthat will affect the total revenues and profits generated by such promotioncampaigns.
Time
12月28日(星期一)14:00-15:30
Speaker
Sheng Bi is a fifth-year Ph.D. candidate in theDepartment of Analytics and Operations at National University of Singapore,advised by Professor Chung Piaw Teo and Long He. Prior to this, she received aBachelor’s degreein Industrial Engineering from Nanjing University. Her research interests arein the area of data-driven optimization, customer choice modeling, revenue andsupply chain management.
Venue
Zoom会议室ID:99930550855
密码:779080