The BYOD app transforms everyday courier operations into a fully connected, data-driven workflow. From parcel validation to proof of delivery and disruption reporting, every action is securely recorded and transmitted to the central platform, ensuring logistics providers remain fully informed. 

When a courier logs in, the application automatically adapts to the configuration of the specific Living Lab deployment. The available features and workflows depend on the operational model of each environment. The BYOD app is designed to support different city deployments with tailored configurations without requiring changes to the core application. 

In the Athens Living Lab, for example, couriers can operate through either Parcels or Stops within the main menu. This flexibility allows the same application to support multiple logistics scenarios without altering the core system. 

Parcel function

In the Parcel function, couriers add parcels by scanning QR codes or by manually entering parcel IDs. For greater efficiency, multiple parcels can be selected at once by scanning code sets or entering a set ID for batch processing. 

Once validated, parcels appear in the current working list, confirming that they are correctly linked to the courier. They remain visible until delivery completion or manual removal or once the delivery is confirmed in the system. A refresh option allows the courier to retrieve the most up-to-date parcel information at any time.  

Selecting a parcel provides access to essential delivery data, including its identification number, status, delivery address, expected delivery date, weight, service type, and associated round. During the delivery process, couriers can register events and update parcel quality directly within the app. 

For proof of delivery, a single parcel is selected, and the receiver signs directly on the device. The signature is securely recorded and immediately reported, ensuring reliable confirmation of delivery and traceability. 

Stop function

Through the main menu, couriers can switch to the Stop function, which provides a structured overview of planned stops and related parcel information grouped by delivery location. Stops can be visualised on an interactive map, offering clear route visibility and improved situational awareness through real-time geolocation.  

Selecting a stop reveals the parcels assigned to that location, allowing couriers to manage grouped deliveries efficiently. If a disruption occurs, the courier can report it directly within the app by selecting the disruption type, adding comments, and automatically sharing their position. This real-time communication supports immediate operational adjustments and proactive issue management.  

BYOD impact

The Green-Log BYOD tool ensures that every action, from parcel validation to signature capture and disruption reporting, is securely transmitted to logistics operators. This continuous flow of information enhances: 

• Transparency 
• Improves coordination 
• Supports data-driven decision-making 

By combining flexibility, live operational visibility, and secure reporting, BYOD strengthens last-mile delivery efficiency while contributing to more sustainable and optimised urban logistics operations across different city environments. 

→ Check our Digital Twins solution

La entrada Bring Your Own Device overview se publicó primero en Mosaic Factor.

Within the AutoMoTIF project, this challenge is addressed through automated shunting as a service, with Mosaic Factor leading the development of the simulation framework. 

From operational bottleneck to coordinated rail operations

Shunting plays a central role in intermodal terminals, linking maritime cargo flows with inland distribution. However, traditional shunting operations are often reactive, fragmented across systems, labour-intensive, and energy-inefficient. 

Shunting as a service reimagines these operations as a digitally orchestrated service platform where autonomous locomotives, yard resources, and scheduling systems operate as an integrated ecosystem. The aim is not simply automation, but service optimisation. 

Simulation driving the transformation

Mosaic Factor’s advanced simulation environment replicates the operational complexity of rail terminals, including train movements, wagon marshalling, yard capacity constraints, container handling cycles, resource allocation, and disruption scenarios. 

Autonomous shunting locomotives are modelled as intelligent agents that dynamically respond to congestion, schedule changes, and infrastructure constraints. 

Through scenario modelling, the simulations evaluate: 

• Reduced shunting time 
• Faster wagon turnaround 
• Lower idle and waiting times 
• Optimised energy consumption 
• Increased yard throughput 
• Improved safety and lower operational costs 

This data-driven approach ensures automation concepts are validated before real-world deployment. 

Shunting as a Service Platform

Shunting as a service introduces a shift in how rail yard operations are structured. Instead of a fixed internal activity, shunting is modelled as a service-oriented platform where capacity is dynamically allocated, operations are digitally coordinated, and performance is continuously monitored. 

This approach supports greater interoperability between terminal operators, rail infrastructure managers, logistics providers, and port authorities, while enabling integration with other automated processes within AutoMoTIF. 

Supporting Smarter Rail Terminals 

To ensure realistic outcomes, Mosaic Factor calibrates simulations using historical operational data, planning inputs, and stress-test scenarios that reflect peak demand and future growth. 

The resulting models provide decision-support tools for infrastructure investment, automation strategies, business models, and regulatory alignment, helping reduce risk and accelerate deployment. 

Strengthening Europe’s Rail Freight Network 

By improving rail efficiency, automated shunting supports broader logistics goals, including: 

• Modal shift from road to rail 
• Reduced terminal congestion 
• Lower emissions 
• Safer working conditions 
• More reliable logistics operations 

Through simulation-driven validation, Mosaic Factor demonstrates how automated shunting can increase throughput, reduce delays, optimise energy use, and enhance safety. 

Automated shunting as a service-oriented platform represents more than a technological upgrade. It introduces a new operational model that strengthens the role of rail in Europe’s transport system while supporting a more efficient and sustainable logistics network. 

→ Check our Digital Twins solution

La entrada Automated Shunting as a Service Platform se publicó primero en Mosaic Factor.

The discussion formed part of the broader work around the CIDAI white paper, created through its think tank with contributions from public and private stakeholders. As participants in this effort, we were pleased to contribute to the roundtable, where Stefano Persi shared practical examples and case studies highlighting the role of AI and trusted data sharing in real logistics environments.

Industry insight

Rail freight volumes remain below the European average and a €5M fee exacerbates the challenge. Transporting goods by rail continues to face operation and structural challenges, even as demand is expected to grow significantly.

In line with the Climate Neutrality 2050 objectives, rail freight volumes are projected to double compared to recent historical averages. Achieving this growth will require not only infrastructure investment, but also better coordination, optimisation, and collaboration across the logistics ecosystem.

From CIMALSA’s perspective, multimodality is a central lever for improving logistics efficiency and sustainability. The optimal model combines rail for medium- and long-distance transport and road transport primarily for first and last-mile operations. AI enables this shift by supporting the reallocation of transport flows from truck to rail, optimising routes, schedules, and capacity usage. This approach can significantly reduce emissions compared to road-only transport, while maintaining operational flexibility.

Data sharing and the role of Data Spaces

A recurring challenge identified is the reluctance of operators and logistics agents to share information. While concerns around privacy are legitimate, they often limit system-wide optimisation. Data spaces were highlighted as a key enabler, providing:

• Secure data exchange and controlled access
• Clear rules on how data is shared and used
• Technical foundation for AI tools to suggest routes, estimate costs, and simulate operational scenarios.

By ensuring data trust, data spaces allow AI to support better decision-making without compromising sensitive business information.

Challenges facing the sector

Five major challenges shape the future of rail freight and logistics:

• Pressure for sustainability and decarbonisation
• Resilience in the face of global and geopolitical crises
• Urban congestion and last-mile regulation
• Digitalisation interoperability, and cybersecurity vulnerabilities
• Organisational readiness and technological transformation

These challenges are closely interconnected and require coordinated responses.

Where AI can deliver tangible value 

AI is already demonstrating measurable benefits in transportation and distribution operations, where it can deliver improvements in efficiency, cost, and sustainability. For Mosaic Factor, this includes automation projects for container loading and unloading at ports.

Despite its potential, barriers hinder AI adoption. These challenges include:

• Lack of an AI strategy within organisations,
• governance issues and data fragmentation,
• lack of quality historical data, difficulty in evaluating the ROI of AI initiatives,
• and a complex and inconsistent regulatory environment. Addressing these barriers requires clearer strategic alignment between technology, operations, and regulation.

Participants aligned on a three-level model for the sector:

1. Digitalisation: basic digitisation and automation, where progress is already visible.
2. Data sharing: secure exchange of data, enabling network visibility.
3. Visibility and success stories: more networks than logistics and gaining visibility of success stories.

Advancing through these levels is essential to unlocking the full potential of AI in logistics.

Key takeaways

Logistics efficiency depends on the effective combination of innovation, sustainability, and regulation. The key question for operators is what value is created by sharing data. AI, when combined with multimodality and trusted data-sharing frameworks, can significantly enhance the efficiency, sustainability, and resilience of rail freight and logistics systems.

During the roundtable, Stefano highlighted three projects Mosaic Factor contributed to:

1. Pioneers: Container Transport Forecast (EU Green Ports Initiative)
2. Port of Antwerp-Bruges: Cargo Flow Predictor
3. Disruptive: Detection & Classification of Logistics Network Disruptions

Together, these projects show the practical impact of AI and data collaboration in real logistics operations, grounding the CIDAI roundtable discussion in tangible solutions.

Click here to read the full white paper from CIDAI.

La entrada The role of AI in Multimodal Logistics and Sustainable Rail Freight se publicó primero en Mosaic Factor.

Moving beyond Traditional RAG

Traditional RAG pipelines embed a query, retrieve context, and generate a response.

Agentic RAG introduces intelligence into the process. By classifying intent before deciding whether to retrieve, call tools, or answer directly, companies report cost reductions of up to 40% and latency improvements of 35%.

Core patterns driving adoption

Industry experts point to three architectural patterns that define Agentic RAG:

• Intent-Based Query Routing: determines whether retrieval is necessary or if a direct answer suffices.
• Tool Orchestration with Error Handling: coordinates APIs, calculators, and databases while managing failures gracefully.
• Continuous Cost & Latency Evaluation: tracks token usage and performance metrics in real time.

These patterns allow systems to decide, adapt, and optimise, a critical requirement for enterprise-scale AI.

Architecture in practice

Agentic RAG systems are typically built on three layers:

• Orchestration Layer: the “decision brain” that routes queries intelligently.
• Execution Layer: handles retrieval, tool calls, and LLM inference.
• Infrastructure Layer: provides vector databases, deployment management, and observability.

Unlike traditional RAG, which always retrieves, Agentic RAG evaluates whether retrieval is even necessary, orchestrating the optimal combination of retrieval, tools, and generation.

Provider flexibility through gateway layers

Another key trend is the rise of gateway abstractions that allow developers to switch seamlessly between providers such as OpenAI, Anthropic, Google, and Bedrock. This approach enables:

• Failover routing when providers experience downtime.
• A/B testing without code changes.
• Cost optimization by directing queries to the most efficient model.
• Freedom from vendor lock-in.

Companies are increasingly adopting unified gateways to balance speed, cost, and reliability across providers.

Conclusion

Agentic RAG is no longer a niche experiment but the blueprint for production AI systems. By combining retrieval with decision-making, orchestration, and observability, the technique is setting new standards for efficiency and adaptability in enterprise AI.

“Production AI isn’t about retrieval alone. It’s about intelligence: knowing when to retrieve, when to call tools, and when to answer directly. Agentic RAG delivers that intelligence”.

La entrada Agentic RAG for AI se publicó primero en Mosaic Factor.

Here are the key trends and developments that stood out.

Digital Twins take center stage

Digital Twin (DT) technology was one of the most prominent themes across exhibitor stands. Cities are increasingly adopting DTs to simulate and manage complex urban systems. The most common applications showcased included:

• Emergency and disaster management
• Human behaviour modelling
• Traffic and parking optimisation
• Energy demand forecasting
• Urban planning, such as identifying areas where new childcare facilities are needed

Several companies also presented the evolution of the Citiverse concept, part of a European initiative that integrates Digital Twins with cybersecurity, IoT, and other advanced technologies (European Commission Citiverse Project). Another highlight was the introduction of 4D Digital Twins, which incorporate the time dimension to enable predictive urban simulations (Nfold ROI).

Smart Cities and visual Language Models

AI innovation was another major focus. NVIDIA unveiled its Visual Language Model (VLM) platform for cities, designed to transform sensor-captured image data into an intelligent “city brain” capable of interpreting current and potential urban scenarios.

Practical applications were demonstrated in Leipzig, where AI-driven DTs are being used to optimise parking spaces and bicycle infrastructure. Meanwhile, the University of Hamburg showcased collaborative, open-source AI projects, emphasizing their interest in joining European-funded initiatives (DCS Intro 2024).

Cybersecurity and Global Engagement

Cybersecurity was a recurring theme throughout the expo, underscoring its critical role in safeguarding smart city infrastructures. Notably, the World Bank was actively involved, reflecting the global importance of secure digital ecosystems.

Our Contribution: Open Innovation Challenges

As part of our participation, we engaged in open innovation challenges, presenting proposals that leverage Large Language Models (LLMs) and Digital Twins for corporate applications. These initiatives demonstrate our commitment to pushing the boundaries of AI and urban technology, ensuring that cities of the future are not only smarter but also more resilient and inclusive.

Our presence at Smart City Expo Barcelona reaffirmed our role as a forward-looking technology provider. By contributing to discussions on Digital Twins, AI, cybersecurity, and open innovation, we continue to shape the future of urban living, driving solutions that make cities more adaptive, efficient, and human-centered.

La entrada Smart City Expo Barcelona se publicó primero en Mosaic Factor.

Let’s break down the key types of AI and how we help you apply them.

AGI: the long-term horizon

General AI, or General Artificial Intelligence (AGI), refers to machines that can perform any intellectual task a human can. It’s flexible, autonomous, and capable of reasoning across domains.

Current Status: AGI is still theoretical. No existing system has achieved true general intelligence.

We think this is a highly compelling topic, and we are monitoring developments with great interest and anticipation.

Generative AI: real-world creativity at scale

Generative AI (or GenAI) is already transforming industries. These models create new content -text, images, code, audio, amongst others- based on learned data patterns.

Use Cases we can deliver for GenAI:

• Automated content generation for specific industries (like healthcare).
• Document summarisation and contract analysis for legal or compliance teams.
• Intelligent chatbots for internal company queries or customer support.
• Code generation and debugging tools for industry-specific developers (like automotive).

Our Solutions: we are capable of building and fine-tuning generative AI models using your proprietary data, ensuring outputs are accurate, brand-aligned, and compliant. Whether you need a custom GPT-style assistant or an image generator for product design, we can make it happen.

Specific AI techniques

Our core focus is enabling businesses to solve precise challenges through advanced AI techniques. It is what we have always done -what we call ‘traditional AI’ – and it has been central to our journey since our founding.

Our Approach: We design, train, and deploy these models with full lifecycle support: from data strategy and infrastructure to governance and performance monitoring.

Why partner with Mosaic Factor?

AI is powerful, but only when applied with precision. We don’t just deliver tools, we deliver transformation.

• Strategic AI consulting and roadmap development
• Custom model design and integration
• Scalable cloud and edge deployment
• Ongoing support, compliance, and optimisation

Whether you are exploring generative AI for creative automation or machine learning for operational efficiency, we help you turn potential into performance.

Ready to explore what AI can do for your business? Contact us to build something extraordinary, together.

La entrada General vs. Generative AI: what they mean for your business se publicó primero en Mosaic Factor.

So we are talking about having open, compliant models based on diversity and ethics, at a European level.

Industry-specific LLMs

Working towards the development of industry-specific language models based on OpenEuroLLM models offers a unique opportunity for companies. These models not only democratise access to high-quality AI technologies but also allow precise customisation to meet the specific needs of each sector.

Key Benefits:

1. Adaptability and Precision: The models can be fine-tuned for specific applications, improving the accuracy and relevance of AI solutions in industrial contexts.
2. Regulatory Compliance: Developed within the European regulatory framework, these models ensure that AI solutions comply with current regulations, reducing legal and ethical risks.
3. Linguistic and Cultural Diversity: The multilingual capability of these models preserves linguistic and cultural diversity, enabling companies to operate effectively in multiple European markets.
4. Transparency and Community: The open nature of the project fosters collaboration and knowledge sharing, creating an active community of developers and users who can contribute to the continuous improvement of the models.

Based on Trustworthy AI

For companies, investing in the development of industry-specific language models based on OpenEuroLLM is not only an innovative strategy but also a way to ensure they are at the forefront of AI technology, complying with European standards and fully leveraging AI capabilities to enhance their competitiveness in the global market.

→ Check our LLMs solution

La entrada Open LLMs for AI transparency se publicó primero en Mosaic Factor.

• Expanding functionalities and ensuring compatibility across components.
• Version 3 of the platform will introduce interactive features, allowing users to adjust optimization parameters directly within the application.
• Version 4 will extend all functionalities to include all Living Labs, ensuring adaptability to diverse urban contexts.

The final version will incorporate the enhanced optimisation module, integrating simulation workflows to create a fully operational platform capable of addressing complex logistical needs. Through continued iteration, stakeholder collaboration, and meticulous testing, the AIMP is on course to deliver a robust and adaptable solution for urban logistics, addressing the needs of Living Labs and showcasing its potential in real-world applications.

The simulation platform

Here you can have a sneak peak on how the AIMP looks like:

→ Check our Digital Twins solution

La entrada Augmented Intelligence Modelling Platform se publicó primero en Mosaic Factor.

To provide an idea, the top-5 algorithms we use more often for predictive models are:

1. Statistical Models: sophisticated statistical models and approaches such as generalised modeling, regularisation, Bayesian Inference, and time series analysis and forecasting which are used to capture intricate dependencies, model uncertainty, and make robust predictions with generalised models based on complex data distributions and latent structures.
2. Machine Learning Algorithms: powerful models to capture complex data relationships with tree-based, kernel-based, ensemble techniques (bagging, boosting, stacking and blending, and voting ensembles). The advanced supervised ML approaches are enhanced with techniques to improve generalisation and interpretability. Reinforcement learning through environment interactions by applying optimisation of policies, value-based learning, and actor-critic methods are designed for (sequential) decision-making.Advanced and tailored unsupervised learning techniques to focus on discovering hidden patterns, creation of segments and groups are developed and used. These techniques include:
   1. clustering,
   2. dimensionality reduction,
   3. and representation learning.
3. Deep Learning techniques: Deep Learning is based on deep neural networks to learn hierarchical representations of data, being key in applications such as natural language processing and image recognition.
4. Neural Networks are advanced models and approaches from deep learning. Representation learning, and attention-based architectures that enable state-of-the-art and also beyond-state-of-the-art with innovation in areas like computer vision, natural language processing, and sequential modelling. The motivation stands for:
   1. improving generalisation,
   2. scalability,
   3. and interpretability through advanced techniques by pushing the boundaries of what a machine can learn.
5. Explainable Artificial Intelligence (XAI techniques): methods aiming to uncover how models with complex dataset and structure make the predictions, providing transparency in decision-making pipelines and processes. Techniques include both model-agnostic and model-specific approaches; they are crucial to understand the rationale behind a model output and a decision.

→ Check our Predictive Model solutions as well as our Trustworthy AI solutions.

La entrada Our top algorithms for predictive modeling se publicó primero en Mosaic Factor.

What are LLMs?

LLMs (Large Language Models), are advanced AI systems capable of understanding and generating various forms of content, including text, code, images, video, and audio. These models are trained on at least one billion parameters (data points), which allow them to grasp language patterns and respond appropriately.

LLMs find applications in natural language processing tasks such as text generation, translation, sentiment analysis, data analysis, question answering, and text summarisation.

Evolution of LLMs

Key milestones include:

• 1966 ELIZA: The first chatbot simulating a psychotherapist.
• 2013 word2vec: Efficient methods for learning word embeddings from raw text.
• 2018 GPT and BERT: Groundbreaking models.
• 2020 GPT-3: A significant leap.
• Late 2021 and 2022: GPT-4 and other advancements.
• Statistical models: Developed to learn patterns from text data.

LLMs vs. NLP

While NLP (Natural Language Processing) models interpret or transform existing text, LLMs excel at generating new, coherent text from scratch.

They can create essays, stories, and even computer code that mimics human writing styles.

Light LLMs

Nowadays, though, there is an increasing importance of smaller models (light LLMs) for specific domain applications.

While the largest models would all be “general purpose”, light LLMs are developed with a specific sector use in mind.

That is:

• Large models use a huge number of parameters, without tuning to a specific use, use a lot of energy, sometimes with questionable reliability, and that provide answers even when they don’t know them.
• Smaller models consider the use that is going to be given to it, refining its responses (fine-tuning) the specific model for a specific use.

Light LLMs benefits

1. Efficiency: Light LLMs require fewer computational resources, making them faster and more cost-effective.
2. Scalability: Companies can deploy light LLMs across various applications without straining infrastructure.
3. Customisation: Light models allow fine-tuning for specific tasks, tailoring them to company needs.
4. Privacy: Smaller models reduce the risk of inadvertently leaking sensitive information.
5. Easier Maintenance: Light LLMs are simpler to manage and update.

To conclude, while both open-source and closed LLMs have their merits, light LLMs offer practical advantages for companies seeking efficient, adaptable solutions. Therefore, you should consider your specific requirements when choosing the right LLM for your organisation.

→ Check our LLMs solution

La entrada What are light LLMs? se publicó primero en Mosaic Factor.