Transport

Data Hub design and implementation

A transport operator had multiplied Data initiatives without results. No centralization, no data culture, aborted projects. We started the subject from scratch: strategy, infrastructure, concrete use cases and team acculturation. At the output: an operational DataHub, 4 dashboards in production and 7 Data Science models deployed.

Problem

The client was coming out of several aborted Data initiatives. Data remained scattered in non-connected systems, no centralized infrastructure existed, and above all: management and businesses had lost confidence in the ability of Data to produce concrete results. The challenge was not only technical: it was necessary to simultaneously build the infrastructure, deliver visible results quickly to restore trust, and lay the foundations for a Data-Driven culture that did not exist in the organization.

Vue rapprochée d’une coupe transversale colorée d’une géode montrant des couches concentriques de minéraux en jaune, marron, rouge et vert.

Solution

What we built

We worked across the entire Data spectrum from strategy to industrialization with a two-stage approach: quick wins to demonstrate value, then structural deployment on all Data subjects.

Step 1 — Audit and strategy. Data Assessment of the organization's Data maturity, inventory of existing technologies, mapping of group databases and impact study on the IS. Definition of the Data strategy, drafting of the specifications for the tools (ETL, hosting such as Snowflake, analysis tools) and benchmark solutions.

Step 2 — Design and construction of the DataHub. Infrastructure architecture on AWS, definition of the Data ecosystem, establishment of the environment. Development and industrialization of ingestion pipelines in Spark and Scala. Establishment of the DevOps ecosystem.

Step 3 — Deployment of BI use cases. Analysis of business needs, data mapping and frameworks, implementation of Power BI and Tableau. Delivery of 4 dashboards in production: scorecard drivers and managers, network attendance, management dashboard for COMEX.

Step 4 — Development and industrialization of 7 use cases. Data Science Automatic incident classification (NLP), predictive maintenance on rolling stock (Time Series), license plate detection and reading (Computer Vision), social media sentiment analysis and satisfaction surveys, and customer churn prediction.

Projects in the same category

See all projects

Supply Chain Optimization Application

A pharmaceutical distribution player had to rethink its entire supply chain: pharmacy assortment, inventory management and delivery channels. Operational research alone was no longer enough. We built the application that transformed its operations.

Social media trend detection

In a market where consumer behaviors change faster than decision cycles, a customer needed to anticipate trends instead of experiencing them. We built the platform that turns social media noise into actionable signals.

Yield of agricultural fields

The climate subsidiary of a French insurance leader needed to predict field yields across Germany to price a new drought insurance offer. Internal data was not enough. We built the predictive models that made the product marketable.

Customer needs analysis platform

A large French group needed to understand the current and future needs of its customers by aggregating a massive volume of consumer data. After a POC of 8 weeks, we industrialized a complete platform deployed on all French entities and is now being extended at the group level.

Skin scoring and analysis

A major player in luxury cosmetics had developed a skin scoring algorithm. Problem: no one could verify what the AI was based on to make its diagnosis. We built the visualization system that makes predictions transparent and deployable in stores and on mobile.

Beaconing detection

A large group's security team needed to identify beaconing signals, regular network traffic sent by potentially compromised machines to servers controlled by attackers. The volume of logs made manual detection impossible. We built the anomaly detection system capable of processing this data on a large scale.