Case Studies

Read stories from our customers:

An agile Enterprise Architecture

Chorégie uses the Obeo SmartEA solution to master its enterprise architecture and to manage strategic transformation plans.

By tracking data to automatically create up-to-date diagrams, Chorégie makes understandable its ecosystem and allows its enterprise architect to be agile by avoiding them useless typing.

Systems Engineering workbench Melody

The Melody ecosystem is a field-proven modeling solution developed by Thales and Obeo. It offers an environment with a high added-value for engineers working on system, software and hardware architectures.

At the center of this ecosystem is a graphical modeling workbench that provides a focus on functional analysis, (complex) architectures definition and early validation.

Railway safety analysis

To improve collaboration between system designer and safety assurance manager, Alstom has created a specific modeling workbench with Sirius.

Supporting a model-based Systems Engineering methodology defined by Alstom, this tool allows safety engineers to graphically describe their analyses (PHA and SHA) by directly requesting the SysML design model of the system. It strongly couples railway signalling system design with risk and safety studies.

Configuration of space systems middlewares

The CNES has defined a model-based approach to generate RTSJ source code, a Java version designed to support both hard and soft real-time application.

In order to easily configure space systems middlewares, Sirius has been used to complement a UML modeling workbench (based on Eclipse) with a DSL dedicated to the communication protocols.

Reuse of satellites software architectures

Sirius has been used to create a modeling workbench for the European Space Agency that supports a generic architecture for on-board satellites applications.

This workbench aims to make software development faster in the context of a reduced schedule, to allow late definition or changes to cope with the various system integration strategies.

Modeling P&C insurance products

The MAIF (a French insurance company) needed a tool providing a structured, centralized and unified vision of P&C (property and casualty) products (e.g. fire or accident) for its insurance business experts.

Obeo has developed a graphical modeling workbench allowing users to directly configure insurance products (server, intranet and new web applications).

Validation of business flows

In the context of social charge returns filed by companies, CNAV has had an incoming flow validator designed (2 millions i.e. several To).

Obeo provided a solution allowing to the business players to describe their checks directly in a federated repository and avoiding lengthy and expensive computer developments.

Agile specification and JavaEE industrialization

The goal of the ministry is to control its application assets from end to end and reduce ownership costs, in particular by favoring Open Source software.

Obeo supported the ministry in creating a collaborative modeling tools and JavaEE code generators deployed across the ministry.

Industrialize the software design

As part of a business application architecture updating process, Pôle Emploi is seeking to industrialize its procedures and its JavaEE development methodology.

Obeo supplied a custom modeling tool and code generators allowing to the public service to reduce delivery times and to ensure greater control of subcontractor relations.

UML adaptation and code generation

CEA DAM (French Atomic Energy Commission, Military Affairs Division) needed to ease the development and the upgradability of its simulation modules.

The approach adopted by Obeo involves describing the modules in UML independent of the execution technology and to generate automatically efficient code for massively parallel computers.

Software migration of 4.5 million lines of code

Following the SUN announcement of the shutdown of the UDS-Forte platform, Obeo developped a solution to migrate the DGFiP business layer application to the Java technology.

All the delivery milestones and quality requirements have been met: the Java code generated fully meets the target architecture for optimum maintenance, and the number of manual adjustments made was less than 4.10-4 (fewer than 4 per 10,000 lines of code).