Optimizing Expected Profit Under Flexible Allocation of Campaign Days and Budget for Management of Shopping Centers: A Machine Learning Approach

Main Article Content

Nora Sharkasi
Jay Rajasekera
Ushio Sumita

Abstract

In retail businesses, a sales campaign is typically organized over segments of consecutive days within a certain period so as to maximize the total sales in that period. It is also common to design a sales campaign in such a way that good-sales-days of the previous year would be designated as sales campaign days, with the expectation that the campaign effect could enhance the potential of good-sales-days further. Contrary to these common practices, Sharkasi, Sumita, and Yoshii (2015) showed that the expected total sales can be increased by reallocating sales campaign days in a more flexible manner. The paper is limited, however, in that the focus was exclusively on the expected total sales without any regard to the expected profit, which is the ultimate objective of any business. Furthermore, the sales campaign budget was totally ignored. The purpose of this paper is to fill this gap by incorporating a constraint for the sales campaign budget, where the expected total sales would be increased as a concave function of the budget increment. Furthermore, the expected profit rather than the total expected sales is optimized. Based on real data collected from a shopping center in Tokyo, it is shown that the expected profit could be increased considerably with less cost under the new optimization approach.

Article Details

Section
Articles