Use Case Intelligence RSS Search use cases Search
Databricks Manufacturing global

Michelin

Data Lake / Data Warehouse Modernization · Supply Chain Optimization

Michelin's centralized on-premises platform lacked the openness and flexibility needed for decentralized teams to access data, tools and technologies at scale and build their own analytics use cases.

Read the original story Explore all use cases
predict stock levels in 15 days' time
scale to hundreds of use cases on the platform
we don't know what 80% of them are, and that's fine

Reported impact

Business outcome

Michelin can predict stock shortages 15 days ahead, scale to hundreds of use cases, let decentralized teams build and share their own data products, and improve collaboration across the organization without the rigidity of the former on-premises platform.

  • Cost Reduction
  • Productivity Improvement
  • Scalability

Solution

Michelin turned the Databricks Data + AI Platform into a self-serve citizen data platform so employees and feature teams could mix local and central data, build their own data products and run analytics without being blocked by a centralized on-premises stack. Data from back-office systems, manufacturing sites and supply-chain processes is unified in a lakehouse where Delta Lake keeps continuously updated datasets available for model training and real-time analysis. Databricks SQL gives analysts a common way to query and visualize results, while isolated cloud workspaces let teams experiment safely and share data products back to the wider organization. One early application used machine learning models trained on historic supply-chain data to predict stock levels 15 days ahead and recommend actions that avoid outages. The same platform now supports hundreds of distributed use cases while preserving governance, collaboration and autonomy for Michelin's decentralized teams.

Data flow

Back-office, manufacturing-site and supply-chain data is unified in the Michelin lakehouse, where Delta Lake keeps the datasets current for machine learning and operational analytics. Teams query and visualize the shared data with Databricks SQL, then publish their own data products and models back to the wider citizen-data-platform community.

Solution architecture

3 components · 3 layers
  1. Compute
    • Databricks Data + AI Platform Serves as Michelin's self-service citizen data platform for decentralized teams and data products.
  2. Storage
    • Delta Lake Keeps unified business and supply-chain data continuously updated for analytics and model training.
  3. Serving
    • Databricks SQL Lets business analysts query and visualize platform data for reporting, forecasting and other use cases.

Architecture clues

  • Built as a self-serve citizen data platform with isolated environments in the cloud.
  • Teams mix local data with central data for a 360-degree understanding of their markets.
  • Uses ML models to predict stock shortages and recommend actions before outages occur.

Evidence from the source

Allowing employees to access Databricks directly through their notebooks to perform their own transformations and analytics has allowed Michelin to realize their vision of building a citizen data platform.
An early use case involved employing AI to predict the shortage of stock.
Using lakehouse means we've been able to scale to hundreds of use cases on the platform
the application was designed to predict stock levels in 15 days' time and recommend actions to prevent outages from occurring in the future.
with real-time capabilities, the data itself was continuously streamed and kept up-to-date in Databricks Delta Lake while business analysts used Databricks SQL to run queries and visualize insights.
Transparency Evidence & methodology 100/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    2 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    5 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Databricks

    Michelin customer story

    First-party
    https://www.databricks.com/customers/michelin
    Source type
    Vendor-published customer story
    Published
    Nov 6, 2023
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Nov 6, 2023 · Databricks · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
Databricks Jan 22, 2026 Dongwon Group Manufacturing South Korea However, their legacy data systems were siloed by business units, limiting their ability to support company-wide AI-based operations. 96% Raw material price forecasting accuracy Data Lake / Data Warehouse ModernizationRisk Management Delta Lake · Unity Catalog Databricks Mar 31, 2025 Nexa Resources Manufacturing Latin America Nexa Resources, a global zinc mining leader, faced challenges with legacy data systems, siloed operations and limited scalability, hindering innovation and decision-making. $370,000 Annual savings from predictive models for zinc filtration and cadmium removal Data Lake / Data Warehouse ModernizationRisk Management Agent Bricks · Databricks SQL · Delta Lake Databricks Mar 15, 2023 Cleanaway Commercial and Professional Services Australia Cleanaway had operational, fleet and customer data spread across disparate systems, making aggregation manual, data quality inconsistent and insight delivery too slow for responsive operational decisions and… 2,800+ Waste collection vehicles impacted by route optimizations Data Lake / Data Warehouse ModernizationSupply Chain Optimization Databricks Data + AI Platform · Delta Lake · MLflow