Google > Case Studies > Tchibo: Optimizing demand forecasts with AI to match customer needs

Tchibo: Optimizing demand forecasts with AI to match customer needs

Customer Company Size

Large Corporate

Region

Europe

Country

Germany

Product

Vertex AI
BigQuery
Google Cloud

Tech Stack

Temporal Fusion Transformer Model
Google Workflows
Microservice Architecture

Implementation Scale

Enterprise-wide Deployment

Impact Metrics

Cost Savings
Customer Satisfaction
Productivity Improvements

Technology Category

Analytics & Modeling - Predictive Analytics
Platform as a Service (PaaS) - Data Management Platforms
Application Infrastructure & Middleware - Event-Driven Application

Applicable Industries

Retail
Consumer Goods

Applicable Functions

Logistics & Transportation
Business Operation

Use Cases

Demand Planning & Forecasting
Predictive Maintenance
Supply Chain Visibility

Services

Cloud Planning, Design & Implementation Services
Data Science Services
System Integration

About The Customer

Tchibo is one of the leading retailers in Europe, with its origins in roasting and selling coffee. It is one of the largest coffee roasters globally and has expanded its business model to include coffee-making equipment and a wide range of non-food items and services, such as communications. Tchibo operates 900 stores and has over 11,300 employees worldwide, with a presence in more than 24,300 retail depots. The company has evolved into one of the biggest e-commerce businesses in Europe, offering a frequently updated range of products and services. Tchibo's operations are supported by a wide logistics distribution network, and it brings in around 3,000 new products annually, with some campaigns planned a year in advance.

The Challenge

Tchibo, a leading retailer in Europe, faced challenges in managing supply and demand for its wide range of non-food items, which include clothing and homeware. The company's business model involves fast-changing weekly sales phases and multi-channel distribution, which requires a robust logistics distribution network. With around 3,000 new products introduced each year and some campaigns planned a year in advance, optimizing logistics processes was critical for cost savings and fulfilling customer service expectations. Tchibo's previous data analytics solution was manually maintained and lacked the forecast quality needed to manage its operations effectively. The company sought a state-of-the-art platform for developing data solutions with machine learning and advanced analytics to improve its demand forecasting capabilities.

The Solution

Tchibo partnered with Google Cloud to build an online demand forecasting service known as DEMON, which accurately predicts demand for the online sales channel. The service uses a temporal fusion transformer model (TFT) on Google Cloud for time series forecasting, relying on an event-driven microservice architecture. The model training and production are separated, and the models, versions, and experiments are centrally managed and easily deployable. DEMON helps Tchibo manage its warehouse supplies and reduce the time spent by operations teams on logistics. The forecast covers 84 days for each item, and the entire process is scheduled and orchestrated by Google Workflows. Forecast results are stored and integrated into subsequent IT systems, such as ERP and allocation, to automate warehouse replenishment. The platform generates over six million optimized predictions a day, ensuring Tchibo has the right level of products to meet customer demand in a timely and cost-effective manner.

Operational Impact

Tchibo's demand forecasting service has led to significant time and cost savings by reducing overstock and handling efforts.
The platform has helped Tchibo anticipate a significant sales increase by reducing stock-outs.
The forecasting model supports data-driven decisions and maintains Tchibo's market-leading position by ensuring the right level of products to meet customer demand.

Quantitative Benefit