AURA - the Business Expert 

Empowers any user by leveraging Enterprise Knowledge & Know-How

LEO - The Engineer

Effectively solve technical challenges in all engineering disciplines

MARIE - The Scientist

Bring Deep Scientific Knowledge to Engineers or Researchers

Knowledge that already exists finally gets used with AURA.

You stop answering the same question twice. People find existing answers before they even finish asking.

Challenge
Engineers design complex mechanical products and engineering workflows in CATIA that require deep domain expertise to navigate and understand intricate product structures and shapes, managing engineering knowledge.

Solution
LEO | Mechanical Engineering reveals the invisible and help anyone to understand a design made by others or rediscover his own design, and manages modification of geometric parameters, relations, and wireframes, and of Product Structures

Benefit
Engineers are up-to-speed in a quicker way to modify and re-use existing designs, and to better understand customer existing assets, protecting Customer asset value.  

Challenge
Designing products and components with future evolution and reuse in mind demands both a deliberate mindset and dedicated time. Yet mechanical teams are often under intense delivery pressure, leaving little room to build on past work. As a result, too many designs are treated as one-off, disposable solutions, useful only once, forcing engineers to repeatedly reinvent the wheel and recreate the same errors. This cycle not only wastes time and talent, but also undermines the scalability and quality of the entire product portfolio.

Solution
Automate the instantiation of any Generative Representation, such as Engineering Templates.

Facilitate Engineering Template retrieval by searching in priority in Data Set up resources.

Benefit
Organizations preserve critical engineering knowledge, accelerate engineering reuse, and ensure consistency across products and programs — turning individual expertise into a scalable, operational enterprise asset rather than a fragile, person-dependent one.

Challenge
Adopting MBSE (Model-Based Systems Engineering) remains challenging for many organizations. Creating and maintaining systems architecture models requires specialized expertise on SysML language and deep familiarity with MBSE methodologies and best practices. 

Solution
LEO | Systems Architecture is a Virtual Companion that acts as an on-demand expert , making MBSE more accessible and accelerating system architecture models development. helps engineers reveal, understand, create and improve MBSE models using natural language. Acting as an systems engineering expert assistant, it leverages industry methodologies such as MagicGrid to guide engineer throughout the systems modeling lifecycle.

Benefit
Make MBSE accessible to more engineers to transition from “document-based” to “model-based” systems engineering. 
Accelerate MBSE adoption by empowering Systems Engineers to create consistent systems architecture, and to increase development efficiency, leveraging built-in industry methodology and natural language to interact with models.

Challenge 

When designing complex systems, there are many parts that repeat - with variations - throughout the design. In construction, this includes many architectural elements such as doors, windows and staircases. Template-based engineering allows designers to quickly create a new element based on a template and modify it for the requirements of this specific part. However, this requires in-depth knowledge of the range of templates available and the parameters to vary, and in a large project, template management is a challenge.

Solution

LEO can manage large collections of templates and search through them to find a specific template requested by the user. It can list all the parameters, their possible values, and their effects. And, once the user has selected the right template, it can instantiate the element automatically, and modify the parameters if the user asks.

Benefits 

LEO reduces the overhead associated with template-based engineering, allowing designers to find the templates they need and work faster. It makes complex templates understandable and can explain how they work in a natural language interface, opening access to the templates and democratizing their use to all team members.

Challenge 

A robot hand represents a complex system with many moving interdependent parts. Managing the complexity of the parts and their interactions can be challenging, especially once the design needs to be analyzed and adapted.

Solution

LEO understands the model and can answer natural language questions about the function of components and the different parameters. It can adjust parameters on request, and make changes to materials and designs.

Benefits 

LEO allows designers to understand complex models better and work faster and more efficiently. Users can get started with an unfamiliar design faster, or find the parameter they need with a simple natural language search.

Challenge 

Significant expertise are required to review the quality and completeness of a SysML model. Issues like missing elements, broken traceability, inconsistent diagrams could go undetected until reviews in later stages. 
When stakeholders requirements are modified or a change need to be implemented, impact analysis can be time-consuming to identify models elements to be updated.

Solution

With the Virtual Companion LEO, users can analyze the systems model in natural language to a get model summary, check model quality and identify issues or gaps and recommend action points.

Benefits 

Using an AI makes model review accessible to more engineers, democratizing access to the SysML model.

Challenge 

In a complex system, there are numerous components all interacting with each other. Each possible interface between parts needs to be defined within the model in order to understand its behavior, and this can be a time-consuming process with a risk of error.

Solution

With the Virtual Companion LEO, users can analyze the systems model in natural language and see a prediction all the mechanical interfaces. LEO can then also help to define the connections.

Benefits 

LEO speeds up the process of identifying interfaces and defining the connections between parts, accelerating design while also reducing the risk of incorrectly defined interfaces.

Challenge 

Simulation results require significant interpretation effort. Extracting meaningful KPIs and understanding system behavior from the results is not straightforward, especially when analyzing multiple scenarios.

Solution

With the Virtual Companion LEO, users can automatically process simulation results to derive KPIs and generate revealing plots. Then, produce a natural language analysis of the system behavior, transforming raw outputs into actionable engineering insights.

Benefits 

LEO speeds up the process of deriving KPIs and generating plots directly from the simulation results. Users can use natural language to analyze system behavior based on simulation results, reducing the time and effort needed to extract KPIs.

Challenge

Assembling parts is a tedious, manual process, made worse when parts lack semantic information — forcing engineers to rely on carry-over methods and manually position and constrain each component rather than working from clear, structured design intent.

Solution

CATIA Generative Assembly predicts mechanical interfaces and automatically generates assemblies from a set of positioned parts. By identifying how components should connect them automatically, it eliminates the need for manual assembly constraints and manual placement.

Benefit

Engineers save significant time and effort by removing the tedious, manual assembly steps — even when parts carry little or no semantic information. Assemblies can be generated faster and more consistently, reducing reliance on manual positioning expertise and freeing engineers to focus on higher-value design decisions.

Challenge

Engineering knowledge is captured across multiple, disconnected documents, making it complex and time-consuming — often taking hours — to review whether rules have been correctly applied. This review typically happens only after the design is already realized, forcing multiple costly iteration loops rather than getting it right the first time.

Solution

CATIA Generative Engineering Rules generates context-aware design rules that guide engineers toward compliant and manufacturable solutions — bringing rule guidance into the design process itself, rather than as a review step after the fact.

Benefit

Engineers achieve compliance by design, with fewer iteration loops and clear, actionable rules available at the point of design — reducing hours of manual rule-checking and catching issues before they become costly downstream reworks.

Challenge
Regulations and certification standards evolve continuously, and keeping specification models aligned with the latest regulatory requirements is difficult when that information lives in static engineering documents (PDFs) rather than in a structured, actionable form — creating risk of certification delays or non-compliance as regulations change.

Solution
Generative Systems Requirements Creation captures the continuous evolution of regulation as an actionable model. It takes engineering documents in PDF form as input and automatically generates structured specification models that are directly usable within the design environment — turning regulatory text into design-ready requirements.

Benefit
Organizations can keep pace with evolving regulations without manual re-authoring of specifications, ensuring certification readiness and design consistency, while significantly reducing the time and effort needed to translate regulatory documents into actionable engineering requirements.

Challenge
Building a SysML model asks three things of an engineer at once: MBSE expertise, fluency in the SysML language itself, and deep familiarity with the methodology plus the company's own modeling rules and best practices.
 

Solution

CATIA Generative Systems Architecture suggest relevant model elements from a set of selected elements (i.e : from Requirements to use case description) as a support to initiate and refine the SysML model and its diagrams (systems, functions, components, ports, parameters). Traceability links between the suggested elements and their source requirements are generated along the way, keeping the specification-to-solution thread as the architecture grows. Suggestions are configurable: the tool can be set to follow company's methodology rules and modeling best practice
 

Benefit
Practitioners who are not SysML specialists can produce a consistent, methodology-compliant architecture, while experienced modelers skip the repetitive scaffolding and spend their time on the design decisions that actually need judgment and creating added values.

Challenge

All manufacturing processes have some degree of tolerance, and the tolerances of different components need to be managed in order to ensure that the system functions reliably when assembled. In a complex system, there will be hundreds of different functional tolerances to take into account, and every single one must be checked and validated. This requires considerable effort and always carries the risk of missing a crucial component. These tolerances must also be clearly communicated to the entire team.
 

Solution

Generative Tolerancing automatically finds and annotates all the tolerances in the system, using the mechanical definitions of the parts and the relationships and interfaces between them. 
 

Benefit
A Generative Experience reduces hours of manual work to seconds, dramatically accelerating the annotation process. 

Challenge

CATIA parts contain a great deal of data about the object in question, such as parameters and contextual information about individual elements. Geometry imported from other sources usually lacks this information, limiting the ability of designers to modify and adapt this imported part.

Solution

Generative Mechanical Part analyzes the geometry and identifies elements such as fillets and chamfers. From the imported "dumb" geometry, CATIA can reconstruct the part while keeping the features in a parameterized form. With assistance from LEO, this can be done automatically with a straightforward natural language interface.

Benefit
Generative Mechanical Part accelerates the design process signficantly, reducing or eliminating the need for the designer to manually recreate imported geometry.