AI Case Study
Granarolo reduces inventory levels by more than 50% and cuts capital and lead time in half using machine learning
Granarolo leverages machine learning technology from ToolsGroup. Using the SO99+ and Trade Promotion Forecasting (TPF) solutions, which use advanced machine learning analytics and deep learning, the company is able to plan demand for perishable products, optimise its inventory and accurately estimate future promotions based on historical data. Granarolo has managed to reduce its inventory levels and delivery time by 50% and increase its average forecast reliability by 5%, resulting in better customer service and fresh products offering.
Consumer Goods And Services
Food Beverage And Drugs
"The Demand Planning process includes collecting information from around 850 merchandisers and its internal operations, including promotions, assortments, and seasonality to define a demand plan which translates dynamically into production and distribution plans.
Granarolo manages this process with ToolsGroup’s SO99+, which is capable of creating a highly reliable demand plan for perishable products. It also identifies possible deviations from the plan in a timely manner, and optimizes the inventory in order to maximize customer service levels, while complying with all logistical constraints.
Moreover, to manage the high pressure promotions and correctly estimate peak demand, Granarolo adopted ToolsGroup’s innovative Trade Promotion Forecasting (TPF) module, which uses machine learning technology to translate historical data into reliable estimates of future promotions. On the basis of past promotions, TPF automatically generates proposals consistent with promotional peaks. The system proposes dynamic safety stock levels that take each product class’ forecast accuracy and store replenishment frequency into account so Granarolo can maintain high service levels in the face of changing demand.
Finally, ToolsGroup’s software is now used for planning and inventory optimization across the strategic, tactical and operational levels of the business. Granarolo is a rapidly growing company in an industry that typically sees no major expansions. Since growth is partly due to acquisitions and geographic expansion, causing frequent changes in the structure of the distribution network, SO99+ proved to be a fundamental tool to support decision-making at the strategic, tactical and operational levels, all with the same base data."
"The dairy market is characterized by short shelf life products and strong promotional pressures. For instance, Granarolo runs thousands of promotions annually, producing 34,000 item–promotion forecasting combinations and causing demand peaks up to 30 times baseline sales. This environment requires optimized inventory management and the ability to provide immediate response times.
Granarolo had grown in recent years, mainly through a number of acquisitions, which resulted in a significant increase in the number of products to manage. Each product had a different set of logistics, requiring Granarolo to reorganize its network and processes. At first, management opted to gradually centralize and rationalize facilities and operations, including transport facilities previously operated autonomously at the local level.
However, Granarolo’s competitive advantage comes from having full control of its supply chain and production network, and a commitment to quality manufacturing and process innovation. The complexity of its planning scenario and the number of products to manage (more than 1200, of which about 200 are milk) called for a complete process overhaul in order to achieve visibility into demand, distribution and production."
Granarolo states the following results:
* Granarolo "brought its average forecast reliability from 80% to 85%, with a peak of about 95% for fresh milk and cream and 88% for yogurt and dessert.
* Inventory levels were reduced by more than 50%, cutting capital and lead time in half.
* By reducing delivery time by 50%, Granarolo also significantly increased product freshness and minimized obsolescence.
* In addition, Granarolo has increased customer service levels, increased sales and reduced transportation costs."
"ToolsGroup’s Trade Promotion Forecasting (TPF) module uses machine learning technology to translate historical data into reliable estimates of future promotions. On the basis of past promotions, TPF automatically generates proposals consistent with promotional peaks. The system proposes dynamic safety stock levels that take each product class’ forecast accuracy and store replenishment frequency into account"
SO99+: "Within our machine learning engine we incorporate deep learning technology that allows our models to “learn” from existing data and accurately identify future demand trends." (ToolsGroup)
historical data of promotions