Trustworthy AI

When using AI models in environments where compliance standards are important, Mosaic Factor can help your company be on top of data governance by applying trustworthy AI solutions.

View solution

Synthetic Data

Synthetic data is artificial data generated from original data using a model trained to reproduce its characteristics and structure.

View solution

Descriptive Models

Descriptive models aim to describe patterns, relationships, and structures within data. They don’t predict future outcomes but provide insights into existing phenomena.

View solution

Predictive Models

Predictive modelling, also known as predictive analytics, is a discipline that uses statistical, mathematical and artificial intelligence techniques to predict future outcomes based on historical data.

View solution

LLMs

At Mosaic Factor, we focus on the creation of domain specific LLMs (or light Large Language Models) for our client organisations.

View solution

Digital Twins

To allow your business to monitor and optimise your assets in real-time Mosaic Factor uses Digital Twins. They can predict failures, detect inefficiencies, and improve decision-making through the use of data.

View solution

Data Enhanced Products

Through different data sources (ie. physical tests) and ML models and usually in combination with our digital twin solutions, our data enhancement solution can learn, predict, and simulate outcomes to provide automatic product configurations that result in product and component improvement during the development process.

View solution

Data As a Service Products

Data as a Service (DaaS) is a cloud-based model that allows companies to access, manage, and analyse data on demand, without the need for extensive on-premise infrastructure.

View solution

Predictive Maintenance

For Predictive maintenance models, we use historical and real-time data to anticipate equipment failures or maintenance needs. By analysing sensor data, maintenance logs, and other relevant information, we can schedule maintenance proactively, reduce downtime, and extend the lifespan of your machinery.

View solution

Demand Cost Forecasting

Our predictive models help businesses forecast demand for products or services. By analysing historical sales data, seasonality, economic factors, and external events we can optimise inventory levels, allocate resources efficiently, and minimise overstock or stockouts.

View solution

Quality Analytics

We identify patterns that correlate with defects or quality issues, allowing your business to take corrective actions early and maintain high-quality standards.

View solution

Inventory Management

We use predictive models to optimise inventory levels by considering factors such as lead time, demand variability, and storage costs.

View solution

Supply Chain Management

We can use historical and real-time data analytics to manage the supply chain, optimise transportation and ensure on-time product delivery.

View solution

Market Understanding

Our descriptive AI models provide valuable insights for decision-making and understanding complex systems of your organisation.

View solution

Pattern Exploration

Our descriptive AI models provide valuable insights for decision-making and understanding complex systems of your organisation.

View solution

Logistics

Logistics

Mosaic Factor’s higher priority in Logistics is sharing key data across different Supply Chain players to optimise performance while managing sustainability by mitigating the impact of these operations.

View industry

Automotive

B:SM Tram Parquímetre

Mosaic Factor’s apply AI solutions in various aspects of the automotive industry, usually by enhancing vehicles and its components during its development.

View industry

Mobility

Mobility

Mosaic Factor’s higher priority in Mobility is to optimise transport systems to people’s mobility while improving overall security and sustainability of transport solutions.

View industry

Corporate Services

Corporate Services

Our machine learning and complex algorithms help organisations manage compliance and customer service to increase the service level of your organization while optimising resolution time for several processes.

View industry

Manufacturing

Manufacturing

Mosaic Factor’s higher priority in Manufacturing is aid our clients decrease costs, increase sustainability while streamlining the production chain.

View industry

Healthcare

Healthcare

Mosaic Factor’s higher priority in Healthcare is making use of data to improve patient care and monitoring in a safe manner to optimise healthcare systems resources and assisting healthcare professionals.

View industry

Open Framework for TwinOps and vehicle specific Digital Twin for Software Defined Evs

Client

Partners

The problem

One of the biggest transformations on history in the automotive industry is being faced. The revolution in the automotive industry about the number of SW components, the number of providers and the continuous Over The Air (OTA) SW updates has arrived while the whole industry is getting ready for adapting to the software-defined vehicle (SDV) concept.

This is a challenge in terms of security and trustworthiness but also an opportunity for electric vehicles (EVs). EVs have three main needs:

    • reducing energy consumption while increasing safety,
    • competing more effectively with traditional vehicles (particularly within the EU sector) in terms of production cost,
    • and enhancing the EV driver experience regarding vehicle charging (availability and power range (forecast).

The energy consumption reduction of each specific function has been the focus of OEMs and Tier suppliers until now, but a new opportunity arises thanks to the computational capacity of clouds and vehicles due to the implementation of High-Performance Computing (HPC) combined with t he digitalization of EVs under the SDV architecture. Nevertheless, state of the art Digital Twin (DT) is still far from the complex reality of EVs core performance, somehow too naive. Besides, a given vehicle, in all its variants and HW and SW versions, is unique and depends on the actual use and its status in terms of health and use of each specific critical components.

Taking into account the unicity of each vehicle, learning from the operational data of a series of vehicles (fleet) and the use of those data and digital models across all EV’s lifecycle (including operations and re-design) in an agile and continuous manner, is the key to unlock the necessary extra step in terms of energy consumption reduction without compromising comfort and safety, but on the contrary enhancing EV driver experience, safety and cybersecurity.

The solution

For solving the problem, it is needed to apply a TwinOps process, by combining in an intertwined manner the specific knowledge of each EV’s health and status, a continuous integration/deployment process and set of tools that allows a continuous and efficient OTA update of EV’s f unctionalities.
Research and development of DTs applied to EVs must be carried out to be one step closer to achieve a specific DT for EV. Exploitation of digitalization is needed to reduce development and validation time and therefore reduce costs to increase competitiveness of EVs and EU automotive sector. All of this based on an automotive level solution (lower cost and high reliability) and considering the impact in terms of driver experience.

The TWIN-LOOP will develop an Open Framework for TwinOps for EVs and a suite of digital tools for continuous improvement of Energy Consumption reduction, Hardware Costs minimization, Driver Experience and Vehicle Resiliency across the 4 stages of vehicle lifecycle. The TwinOps concept is the combination of Digital Twins over a continuous integration/deployment production cycle (DevSecOps) and it leverages other sources of truth (e.g., CAD, Physics) to improve SW Verification and Validation (V&V), using precise models instead of (naive) abstractions. The specificity of each EV is taken in account (MyEV concept) in order to improve each stage, from design to validation in an infinite loop.

MOSAIC FACTOR is leading this project and we will contribute to the definition of the uses cases for testing and validating the tools developed in TWIN-LOOP, to the adaptation of the DevSecOps methodology for software-defined EVs, and to the definition of the functional and non-functional requirements on the components to be developed in this project.

We will also contribute to the development of data-driven models for dynamic management of the EV and lead the development of MyEV Digital Twin Modules for components monitoring and power range forecasting and will contribute to the design optimization based on vehicle and fleet data.

We will also lead the definition and the implementation of the Open Framework for MyEV Digital Twin and will participate in the integration of MyEV and Extended EV APPS integration, as well as participate in the driver profiling and user engagement activities

Data

Not available yet.

Results

Not available yet.

Do you have any questions?

We are always ready to help you and answer your questions. 





    *Fields marked with an asterisk are mandatory