Building Better Decisions Through Simulation and Context

Quantexa recently announced Q Labs, an innovation hub allowing customers, partners and Quantexans to collaborate on innovation. One exciting project is Simulation Lab, helping businesses model real-world scenarios - simulated realities - to inform critical decision making.

Simulations are key to many organizations’ toolkits for making optimal decisions. Combining simulations with domain models rooted in an organization’s platform architecture allows evaluation of “what if” scenarios - like changes in market conditions, competitive moves, or operational disruptions without real-world consequences.

Exploring how simulation technologies evolved, and how they impacted business and society, particularly in the run-up to, and solutions for, the Global Financial Crisis (GFC) of 2008, can give us context for the use cases of current iterations. Through this lens, we can explore how simulation continues to evolve in the AI age, and why Q Labs is actively engaged.

5 Historic moments of simulation and computing

Computing’s history is intertwined with simulation.

1. The birth of computing was coupled with simulation
   Early computers like the ENIAC (1940s) were developed to perform military ballistic trajectory simulations. Alan Turing’s work on codebreaking at Bletchley Park involved computational simulations to test cryptographic hypotheses.

2. Monte Carlo Simulation
   As computing power grew, the Monte Carlo method (1940s–50s) emerged to aid complex systems modelling. Developed at Los Alamos for nuclear research, Monte Carlo probabilistic simulation was an early high-performance computing (HPC) application. It would soon power weather forecasting, computational finance, engineering, and physics particle acceleration as compute was commoditized.

3. The 20^th century expansion of simulation
   From the 1960s, NASA simulated spacecraft behavior under different conditions, enhanced by Finite Element Analysis (FEA) and Computational Fluid Dynamics (CFD), allowing engineers to simulate structures and aerodynamics before physical prototyping. Elsewhere, structured simulation drove enterprise technology foundations through operations research and economic forecasting.

4. From simulation to code in the late 20^th century

   In the 1990s, Complex Event Processing (CEP) and Model-Based Design (MBD) connected simulation to production, allowing engineers to simulate control systems, signal processing and state machines and then automatically generate embedded code. This helped drive feature proliferation and improved reliability in automotive, aerospace, electronics, communications and industrial hardware.

5. Simulation and synthetic data in AI and machine learning
   Modern artificial intelligence and machine learning uses simulated environments to train models. Reinforcement learning, for example, uses simulated worlds (e.g., OpenAI’s Gym) to learn without incurring real-world risks. Autonomous vehicles and robotics industries apply synthetic data and digital twin simulations for development.

Simulation applicability across industry:

Simulation and the Global Financial Crisis

In financial services, Monte Carlo simulations, generating statistically thousands of potential future market, economic and risk scenarios, drives risk management, derivatives pricing, insurance liabilities projection, capital markets trade simulation, and investment strategies. It is part of the lifeblood of modern computational finance.

However, its misapplication adversely contributed to the 2008 global financial crisis via the misuse of the Gaussian copula function. This mathematical formula assumed that defaults across different assets followed a normal distribution, thus correlated in statistically predictable ways. Financial institutions such as banks and rating agencies used this method to underpin Monte Carlo simulations in then popular complex financial products like collateralized debt obligations (CDOs) and mortgage-backed securities (MBS).

However, the models failed to account for real-world dependencies and relationships between mortgage defaults, defaulters and derivative instruments across the financial system. As house prices fell in the US, it triggered mortgage defaults like a fast-falling house of cards, then contaminating the financial system as CDO and MBS valuations collapsed. Banks failed as global credit froze, while debt infused populations that could least afford it. Our blind reliance on simplified quantitative models, which obfuscated real world context, failed these “Black Swan” events.

Simulation and the stress testing fix

Post-GFC, risk management was transformed. Stress testing became mandated by regulators in both banking and insurance, for example in Basel III, IFRS 9 and Solvency II regulations, to assess resilience under extreme economic scenarios. It brought some realism to stress testing, not just the highest profile regulatory capital-at-risk measures, but transformed the capability and impact of credit, counterparty and operational risk disciplines.

As the world evolves, so too does simulation through agent-based models, AI and behavioral simulations, including those driven by graphs, and, in the longer term, quantum computing.  It is critically important we learn the lessons of the past and adapt our approaches to avoid potentially catastrophic scenarios and ensure people can use the financial system without fear. One way to mitigate against such risks is to evolve our use of simulation, and that financial services firms can - and should - incorporate human decision-making, context, relationships and adaptive strategies into simulation.

Q Labs is bringing contextual innovation to simulation

The critical importance of context & graphs in simulation

In all industries, AI innovation is encouraging organizations to test, audit, and refine AI models in controlled simulated environments, for example in:

• Bias and risk assessment – Simulating under different conditions helps identify biases, unintended consequences, and compliance risks.

• Explainability and trust – Simulating with synthetic or real-world-inspired data can improve model transparency.

• Robustness and security - simulated adversarial testing exposes vulnerabilities and empowers resilience.

• Regulatory stress testing - Running scenarios in some regions to ensure AI-driven decisions align with evolving legal frameworks, e.g. the EU AI Act.

Graph technologies like those at Quantexa help add context, explainability, structure and search potential to simulation, amplifying the importance of relationships, not just in prediction but in discovery. They can help replay and simulate trajectories of past events to refine optimal decisions, for example using evolutionary search techniques such as Monte Carlo Tree method. For example:

• Contextual data for better models – Knowledge graphs organize relationships between entities (e.g., people, organizations, events), enabling simulations to mirror real-world complexity.

• Dynamic scenario generation – By linking diverse data sources, graphs create realistic, evolving simulation environments for predictive models, risk assessment, and AI training.

• Causal inference & explainability – Unlike black-box models, simulations powered by graphs explain why specific outcomes occur by tracing relationships and dependencies.

• Adaptive & real-time simulations – Graphs support dynamic updates, allowing simulations to evolve as new information becomes available.

• Digital twins – Graphs offer structured inputs for digital twins, i.e. virtual environments.

When integrating knowledge graphs with simulation, organizations can improve decision intelligence with grounded knowledge, interpretability, and adaptative model behavior. This is why Quantexa, through Q Labs, is exploring context-driven simulation technologies alongside its entity-centric graph technologies, to deliver impactful and transparent decision-making and AI.

Engage with Q Labs and Simulation Lab today.