F&B Retailing : Replenishment Intelligence

Notably, food retailing is one of the fastest-changing industries, standing in stark contrast to its traditionally conservative nature. The food industry generally grows in direct proportion to population growth. However, food retailers are under unprecedented pressure to adapt due to factors such as shifts in household composition, trend-driven food culture, advanced logistics systems like cold chains and online platforms, as well as climate change and global supply chain disruptions. 

Due to short expiration dates or best-before dates, inventory management for food is of unparalleled importance compared to other manufacturing or retail industries. As a result, demand forecasting has become one of the most popular topics within the AI sector. Consumption patterns vary significantly depending on categories (fresh produce, processed foods, snacks, beverages, alcohol, etc.), weather, seasons, regions, and trends. The countless influencing factors and their complexities highlight the critical importance of adopting AI.

Generally, based on sales patterns, approaches to replenishment are applied differently. For steadily selling products, a dedicated team regularly handles ordering. AI contributes well in these categories due to the abundance of data and clear patterns. The goal of these categories is a fully automated ordering system that considers storage methods (room temperature, refrigerated, frozen), lead time, inventory level, and sales trends. There are some challenges as the roles, interests, responsibilities vary among various stakeholders, such as headquarters, franchise, logistic centres and production teams. 

In contrast, for the new products or promotion sales, a completely different approach is required. Despite efforts to AI implementation, still this category has remained primarily within the domain of human experts. While steadily selling products make up the majority of consumption, new product release is highly trend-dependent. It is not surprising that the growth of the food manufacturers or retailers relie on social media, and public media. The main goal in this category is simple: increase sales. Nevertheless, There are numerous factors that cannot be solely relied upon data, such as external events, promotion plans, and incentives for franchises. The merchandise experts must delicately manage the comprehensive replenishment process by their expertise, knowledge, experience, and objective data analysis. 

DEIN Station's Replenishment Intelligence can enhance both routine/regular orders and new product/promotional orders. While AI cannot support rapid market changes, psychological factors, and experts' intuitive decision-making, Decision Intelligence overcomes these challenges through the interaction of AI and humans. AI focuses on predicting localized issues, whereas Decision Intelligence incorporates all decision-making elements and variables into its process. By utilizing Replenishment Intelligence, merchandisers can integratively perform data analysis, simulation, optimization, and monitoring based on their expertise.

The ultimate goal of Replenishment Intelligence is to optimize the relationship between overstock and shortage to improve profitability. It’s easy to understand that excess inventory leads to losses from spoilage and storage costs. However, understanding the dynamics of the distribution market reveals how fatal shortage can be. Overstock and shortage have a contradictory relationship where improvement in one aspect can deteriorate the other. 

"Balancing in a constantly changing environment"

—how can food distributors overcome this challenge?

Experience a new decisioning framework with DEIN Stations, leveraging Replenishment Intelligence.

Business KPIs 

• Overstock Amount

• Disposal Rate

• Shortage Amount

• Shortage Rate

• Business Loss

Challenges 

• Making balanced decisions among contradictory KPIs

• Reordering for new products 

• Simulation the effects of promotions and external events 

• Optimizing the orders by SKUs and stores 

• Automating the ordering process

Decision subjects 

• Order Amount (by SKUs & Stores)

• Order Timing (by SKUs & Stores)

• Promotions

• Incentives

• External Events