DEMAND FORECASTING AND MARKDOWN OPTIMIZATION IN RETAIL INDUSTRY
Tarih ve Saat
7 Haziran 2018 - 15:00
YerMühendislik Binası A-511
Title: DEMAND FORECASTING AND MARKDOWN OPTIMIZATION IN RETAIL INDUSTRY
Abstract: Markdown pricing is a permanent price reduction which is commonly applied to short life cycle products (SLCPs), such as, fast fashion and consumer electronics (cell phones, digital cameras, personal computers, etc.), which have a significant place in the retail industry. This study aims to develop a multi-product decision support system for markdown pricing of SLCPs which has two stages: demand estimation and markdown optimization. In the first stage, we propose an adaptive prediction model based on Deep Learning. Model parameters are updated with the last demand information to estimate customer response to price changes. In the second stage, we are going to solve the revenue optimization problem by employing Dynamic Programming (DP). The latter takes the output of the first stage as an input and gives the answers of following questions: i) the product to be marked down ii) when to mark it, and iii) the markdown amount. Such a decision support system may play a significant role in retailers’ strategic decisions about markdown pricing. All in all, a successfully implemented system may provide important benefits in terms of higher revenues and reduced excess inventory at the end of the sales season.
Biography: Nur Gülcan graduated from Mathematics Dept., Yeditepe University in 2010. After the completion of her Bachelor Degree in Mathematics, she continues her career at Yeditepe University as a PhD student in Systems Engineering starting on October, 2010.