Research & Projects

Overview

I designed and developed a digital twin prototype to explore how AI-driven simulation can support decision-making in complex systems.

The project focused on creating a flexible, interactive model that allows users to test scenarios, understand system behaviour, and explore potential outcomes in a controlled environment.

The Problem

Organisations operating in complex environments—such as financial services or healthcare—often face:

• High levels of uncertainty

• Interdependent systems that are difficult to model

• Limited ability to test decisions before implementation

Traditional tools provide static analysis, but lack the ability to simulate dynamic, evolving scenarios.

The Goal

To prototype a system that could:

• Represent a real-world system as a dynamic digital model

• Allow users to simulate different scenarios

• Provide insight into how changes propagate through the system

Approach

1. System Design

Defined the structure of the digital twin, including:

• Key entities and relationships

• Inputs, outputs, and feedback loops

• Areas of uncertainty and variability

2. AI Integration

Explored how generative AI could:

• Simulate possible system states

• Generate scenario variations

• Support exploratory analysis

3. Prototyping

Built an interactive prototype to:

• Test core concepts quickly

• Iterate on system behaviour

• Validate usability and usefulness

Solution

The result was a working digital twin prototype that:

• Models a simplified version of a complex system

• Enables scenario-based exploration

• Demonstrates how AI can augment simulation and decision-making

The prototype acts as a foundation for more advanced implementations, including real-time data integration and domain-specific modelling.

Outcomes & Insights

• Demonstrated the potential of digital twins beyond engineering into knowledge-based domains

• Identified key challenges in modelling uncertainty and system complexity

• Validated the value of rapid prototyping when working with emerging technologies

Key Takeaways

• Digital twins can be a powerful tool for decision support, not just monitoring

• Generative AI enables more flexible and exploratory simulation

• Early-stage prototypes are critical for understanding real-world applicability

Next Steps

Future iterations could include:

• Integration with real-world data sources

• More sophisticated simulation models

• Domain-specific applications (e.g. financial systems, healthcare scenarios)

Interested in digital twins or AI-driven simulation?
Let’s discuss ideas or potential collaborations.