Skip to main content

Reducing stock outs with fully automated Demand Prediction and Inventory Optimization

demand-prediction-banner

Stockouts and inventory pile ups are a massive problem across the globe for most retailers and consumer brands. Studies show that companies are losing up to 12% of their annual revenues because of inventory issues. Stockouts and wastage can be reduced by having agile supply chains that can react quickly to any changes on demand or supply side. The key to creating agile supply chains is to have accurate demand predictions that are refreshed frequently, incorporate impact of business/external factors such as pricing changes, and constantly align these predictions with the supply side constraints to provide optimal recommendations on inventory levels that should be maintained in warehouses and stores to keep up with the ever-changing consumer demand.

Business Vertical:

Grocery Chain

Main Challenge:

Reduce stockouts and improve product availability

Region:

India

Size of the firm:

The client is a mid-sized grocery chain with 200+ stores and annual revenue turnover of over $200M

demand-prediction-retail

Challenge

The client is a retail chain that sells consumer packaged goods, beauty, and wellness products. They have over 200 stores selling around 50K SKUs from each store. Just like many of their contemporaries, they are perpetually facing the issue of stockouts which in turn translates to millions of dollars of lost revenues. Their entire demand and inventory planning process is manual, biased by gut feel and fully spreadsheet based. Their demand planning process uses an approach based on average daily orders and with static min-max days norms. Their store replenishment plans are also based on backward looking daily averages. 

The daily average methods are highly influenced by stock-outs, where past stock outs lead to lower average sales leading to less-than-optimal replenishment quantities, which can lead to further stock-outs. On top of this, they have little visibility into the performance of their suppliers as they deal with a wide variety of products and multiple suppliers for each of the products. They tend to go with their default suppliers for simplicity rather than the ones that give best lead times/fill rates/prices. As a result, they are not able to quickly adapt to their ever-changing consumer demand preferences, align the right suppliers that can meet their requirement and replenish the stores with the right product quantities at the right time.

Solutions Proposed

  • Automate procurement plans powered by accurate SKU + Warehouse + Store level demand predictions
  • Continuously score suppliers based on their performance metrics
  • Optimize replenishment from warehouse to stores on future looking demand than backward looking averages
  • Surface early warning on stock outs for proactive procurement/replenishment

demand-prediction-solution

What Qentelli Did

After a careful assessment of the client’s status-quo and the target state of supply chain management, we have evaluated and selected a product that the best fit for the client and also gives us the ability to fine tune the algorithm and continuously monitor the feedback. The deployed solution is intended to bring all interrelated functions tied to sales, demand, pricing, inventory, and purchase orders all on one platform to create a system of intelligence that can enable seamless flow of information across the entire supply chain. 

Picking an AI/ML solution is an ergonomic decision considering the demands of the project and ease of modifying the algorithms to suit the changing needs of the business. Here are some of the unique aspects of our approach:

Omnichannel Demand Sensing - The state-of-the art AI/ML engine deployed in client’s system provides accurate bottom-up predictions for each SKU + Warehouse + Store combination. This considers the impact of several demand drivers such as factors such pricing, promotions, and holidays. Our demand sensing engine also tries to minimize the impact of historical stock-outs and anomalies which leads to more robust predictions of future demand.

Automated Procurement Planning - This functionality combines SKU + Warehouse demand predictions with inventory and purchase order data, dynamically calculates safety stock based on historical errors and determines the right quantities to order given the future demand and current inventory runway.

Best Buy Supplier Scoring - The supplier scoring mechanism that is built into the AI engine continuously tracks supplier performance based on lead times, fill rates and pricing and recommends the best supplier for each SKU for every planning cycle.

Optimized Replenishment - The deployed solution is built to be able to scale understanding the impact of seasonality, holidays, neighborhood effects for each SKU in each store, warehouse to store replenishment quantities and safety stocks are determined using future looking daily demand predictions than using traditional min-max based systems.

Outcomes

  • 27%

    Lesser under/overstocking

  • $500k

    monthly savings on working capital

  • 200%

    Lesser man hours spent in planning

With end-to-end automation powered by advanced AI algorithms, our solution has been able to reduce the client’s understocking and overstocking occurrences by around 27% which translates to savings of $500k per month (10% reduction) on the working capital required. In addition to this, the entire planning process has been cut down from several days to just a few hours which is a major productivity boost to the planning team.

Talk to an expert

Advance Your Business with Our Digital
Innovation Expertise

Schedule a Consultation with Our Experts

Demo

Casestudy-Tittle
Reducing stock outs with fully automated Demand Prediction and Inventory Optimization
casestudy-insight-image
Reducing stock outs with fully automated Demand Prediction and Inventory Optimization