Anael Engel

Labor hours are a crucial factor for store performance, as they affect customer satisfaction and sales revenue. However, determining the optimal number of cashiers for each hour is challenging, as rarely do store owners (or the Data Scientists who try to help them) have the data needed to achieve that. In NCR Analytics, we benefit from having a vast amount of data on sales and cashier performance in retail. In this project, we propose a novel model to evaluate and optimize the labor hours for both Point of Sales (POS) and Self-Checkout (SCO) registers, using commerce and cashier performance data. Our model leverages the net time spent on sales, which is the actual time that cashiers (or SCO) are engaged in transactions, and the traffic, the number of items sold per hour. Modeling the relationship between these variables allows us to identify when the labor hours were excessive, insufficient, or adequate. Our model is robust, interpretable, and applicable to any store. It is efficient in terms of storage and running time and can generate results on the fly. It can help store owners test how many labor hours they could save by adding more SCO lanes, or how much it would cost them if they had none. We demonstrate the effectiveness of our model using real data from a retail chain and show how it can help store owners reduce labor costs and improve customer service.