|
Activity Number:
|
156
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 4, 2003 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Business & Economics Statistics Section
|
| Abstract - #301157 |
|
Title:
|
Developing a Demand-Based Approach to Optimizing Retail Prices and Promotions
|
|
Author(s):
|
Suzanne N. Valentine*+
|
|
Companies:
|
DemandTec, Inc.
|
|
Address:
|
1 Circle Star Way, San Carlos, CA, 94070-6236,
|
|
Keywords:
|
econometrics ; software ; retail merchandising ; data mining ; demand-based management ; optimization
|
|
Abstract:
|
Retailers capture impressive amounts of sales-related data, hoping this information will provide better insight to pricing and promotion decisions. When harnessed, these insights have proven to be extraordinarily valuable, but analytical efforts to date have met with marginal success. Several challenges arise in building cost-effective and reliable decision support solutions for price and promotion optimization: Accurate demand model estimation for the 45 million+ store-product combinations existing in a typical national grocer; Validation for notoriously messy retail data; Sparse sales data for slow moving products; Important but highly correlated promotions; Missing records (may correspond to zero demand or out-of-stock conditions); Constantly changing product mix. Further challenges arise in design and implementation of large-scale nonlinear optimization algorithms that deliver optimal prices and promotion schedules quickly under various constraints representing a retailer's business strategy. This talk outlines a response modeling approach that mitigates the issues mentioned above. An approach to the nonlinear optimization will also be discussed.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2003 program |
