DEMAND LEARNING AND DYNAMIC PRICING FOR PERISHABLE PRODUCTS

Abstract

Dynamic Pricing, the concept of setting variable prices to increase revenues and profits, has been traditionally utilized by the travel, hospitality and entertainment industry. However, the practice is not as common in the fresh produce industry, where the inventory is perishable and demand is uncertain. In many dynamic pricing studies, although much work has gone into deriving optimal prices based on initial estimates of demand, there has not been much focus on calibrating the estimate of demand or other relevant purchase considerations such as quality, over the selling season. This paper attempts to develop a model that allows fresh produce retailers to derive an optimal price to reduce end-of-day wastage and maximise revenues. Demand and quality estimates are recalibrated in the afternoon after morning observations and subsequently, the optimal price is revised. This will allow retailers to remove the variability of human guesswork in setting prices and in its place, implement a systematic method of setting prices. To compare the model’s effectiveness, its performance is compared with the status quo method of static pricing.