Demand-based on-shelf availability - Considering intraday seasonality in shopper demand for retail shelf replenishment and inventory control

Balancing supply and demand is at the core of retail management. In the retail setting, oversupply manifests in inventory and cost, undersupply in out of stocks. This research proposal intends to add to retailing research on forecasting shopper demand and deriving optimum order and replenishment policies on store level.
Research estimates the rate of out of stocks - defined here as any occasion when a shopper cannot find the desired product on the expected place of the shelves at a retail store - to be approximately 8%, with little improvement over the past years despite growing management attention and improved store ordering practices. Store ordering practices and replenishment processes were identified as major root-causes for out of stocks.
Surprisingly little attention has been devoted to understand if all out of stocks occur or matter the same. Some researchers have argued that substitution lessens the relevance of out of stocks. Others found that shopper respond differently to out-of-stocks depending on factors like the kind of shopping trip, or the shopper's brand or store loyalty. Economically, out-of-stocks matter most when they occur in times of high demand. Research still has to investigate whether demand per SKU varies in consistent patterns over the course of a day. Are there recurring periods within a day when a certain SKU is sold in high- versus low-volume, distinguishing periods when on shelf availability is highly or little important? This question is particularly relevant for (very) fast moving SKUs requiring multiple replenishments per day and being highly prone to out of stocks or quickly depleted inventory.
Current approaches of on-shelf availability management do not refine their assessment beyond the aggregated daily demand per stock keeping units (SKU). Today's sales forecasting approaches and replenishment concepts in retailing aim at ensuring product availability at any time of the day, assuming that shoppers potentially request any product at any time.
By questioning today's predominant assumption in retail operations management of stationary demand per SKU over a full day, this research proposal contributes to research by developing differentiated replenishment approaches focusing on circadian demand variations over the course of a day (time).