Artificial Intelligence Drives More Efficient Goods Handling in Retail
Store staff is an essential sales and cost driver in physical retail. In grocery, store labor is the single biggest operating cost, amounting to around 14% of revenue. Therefore, it is always good business for retailers to improve work efficiency in their stores. As retailers struggle to find store workers and see wages increase, in-store productivity is more topical than ever.
Store replenishment has a direct impact on operational efficiency in stores: poorly planned store replenishment leads to inefficient shelf stocking in the stores. In the worst cases, store workers spend a lot of time moving pallets around the store while stocking shelves and have to make frequent trips to and from the backroom storage area to fetch products.
In this white paper, we introduce the concept of main replenishment days and show how their application substantially improves in-store productivity. Moreover, we demonstrate how the use of AI (artificial intelligence) to optimize main replenishment days enables more efficient inventory flows in retail distribution.
Main Replenishment Days for More Efficient Store Operations
Most large retailers replenish their stores every day because fast moving products and fresh items demand frequent deliveries. Also, the delivery volume from the retailers’ distribution centers to the stores are large enough to warrant daily shipments.
Most replenishment systems base their planning on these distribution schedules. If daily deliveries are available, most systems will set safety stocks and estimate order needs based on next-day delivery.
Yet, if all replenishment opportunities are used for all products without discretion, two problems follow:
1. Pallets or roll cages delivered to the stores carry a random mix of products representing many different product categories located in different parts of the store. This means store workers need to spend a lot of time moving pallets around the store when stocking shelves (see Figure 1).
2. The delivery volume to the stores reflects the stores’ daily variation in sales. Often, there are delivery peaks towards the end of the week in anticipation of weekend demand. This leads to fluctuating capacity needs in both distribution and stores, which, in turn, increases costs.
Figure 1: When pallets contain a variety of products, store personnel spend a lot of time moving from aisle to aisle while shelving products. The use of main replenishment days based on a store’s floor plan significantly increases in-store replenishment efficiency. (Illustration from Winning the Food Fight: Best Practices for Managing Grocery Retail Supply Chains.)
The best practice is to apply main replenishment days to consolidate replenishment by concentrating the replenishment of products displayed in the same area of a store, such as a specific aisle, to certain weekdays. Safety stocks and order quantities are then calculated based on these main replenishment days. In the case of unexpected demand spikes, the replenishment system will place additional orders to avoid stock-outs, ensuring the highest possible on-shelf availability.
In practice, this means that not all products are ordered every day — for example, detergents. Instead, fast-moving detergents would be replenished on Mondays and Thursdays, while slow-moving detergents would be replenished only on Mondays. The other available replenishment days would be used only when there is a risk of stock-outs.
Main replenishment days are applicable to all but the shortest shelf-life products. In addition, the products need to have sufficient shelf space to accommodate at least a couple of days’ sales in order to avoid unnecessary backroom stock. In grocery, main replenishment days can typically be applied to all but the fastest moving or bulkiest center store products.
The use of main replenishment days substantially increases the efficiency of shelf stocking in stores without hampering on-shelf availability. We have seen reductions of 20% in time spent on stocking shelves following the introduction of main replenishment days.
Artificial Intelligence Optimizes Inventory Flows
Besides more consolidated deliveries, main replenishment days enable the leveling out of inventory flows. This is especially true when harnessing artificial intelligence to optimize main replenishment days.
RELEX’s AI-based optimization of main replenishment days uses a particle swarm algorithm for multi-objective optimization. The optimization prioritizes objectives based on customer-specific business targets. In some cases, the main priority can be to achieve as smooth a goods flow over the week as possible (see Figure 2). In other cases, a smoother goods flow needs to be combined with lower volumes during the weekend, when labor is more expensive (see Figure 3). In addition, the optimization minimizes the number of replenishment days and shelf breaches, i.e. occasions when deliveries do not fit straight onto the shelf.
The optimization is done per store to find the best main replenishment days for groups of products displayed in the same part of a store. It considers the products’ shelf-life, shelf space and sales patterns. In addition, constraints such as the minimum and a maximum number of replenishment days and the available distribution schedules are respected.
Figure 2: This graph shows the inbound goods flow to a large grocery store before and after main replenishment days had been optimized using AI. As can be seen, the goods flow following optimization is much smoother.
Figure 3. This Nordic retailer wanted to both smooth volumes and reduce volume during weekends when store work is more expensive. As seen in the graph, the AI-optimized main replenishment days enable the retailer to meet these objectives.
Compared to setting main replenishment days based on rules defined by replenishment planners, AI-driven optimization delivers clear additional benefits:
- A much smoother inbound goods flow to stores. This allows for more level workloads and more predictable work shifts in stores, increased cost-effectiveness and significantly fewer capacity issues. In addition, a level base volume makes replenishment planning around major holidays easier.
- As artificial intelligence chooses the replenishment days more accurately, stores receive more of the same and similar products in one go. This makes in-store replenishment more efficient without sacrificing availability or straight-to-shelf deliveries.
- Optimization further decreases the number of store order lines. This means lower picking costs in the distribution centers as well as reduced handling costs in stores.
Retail is undergoing a profound transformation, and only time will tell who the winners will be. In any case, it is evident that retailers can no longer afford to sustain inefficient operations. The ability to keep operational costs in check is essential for profitability and even survival. Applying pragmatic AI to optimize replenishment days and operations is an important means to this end.
Want to know when similar resources are published?
Subscribe to receive a monthly digest of our most valuable resources like blog posts, whitepapers and guides.
- Decision Science and Pragmatic AI in Retail
- Pushing the Boundaries of Data Processing Requires Specialization
- Considering Cannibalization and Halo Effects to Improve Demand Forecasts
- Towards a Weather-proactive Supply Chain
- Fresh Forecasting & Replenishment: Running an Efficient Omnichannel Grocery Retail Operation