Case Study: Enhancing Humanitarian Response with Flee and FabFlee – A Partnership with Save the Children

Introduction to Flee, FabFlee, and the SEAVEA Toolkit's Role

Flee, an agent-based model developed by researchers at Brunel University of London, is specifically designed to simulate and predict human displacement in response to conflict and disaster driven crises. By modelling individual agents (people) and their interactions with the environment, Flee generates detailed insights into migration patterns, population movements, and their potential impact on host communities.

To enhance Flee's capabilities for large-scale simulations and high-performance computing, the FabFlee plugin was developed. FabFlee integrates with FabSim3, a powerful automation toolkit that simplifies the execution of complex computational workflows across diverse infrastructures, from local machines to supercomputers. This integration allows researchers and humanitarian organisations to run Flee simulations efficiently and at scale, generating robust data for analysis.

Flee's development and its extensive modelling and simulation work for Save the Children have substantially benefited from the SEAVEA toolkit (SEAVEAtk). The SEAVEAtk provides advanced computational tools and methodologies, particularly through its FabSim3, EasyVVUQ, and QCJ Pilot Job components. At its core, the SEAVEA toolkit aims to bridge the gap between cutting-edge research and practical application, empowering organisations to make more informed decisions in complex global challenges, especially within humanitarian and societal contexts.

The Challenge: Predicting Displacement for Effective Humanitarian Aid

Humanitarian crises, whether caused by conflict, natural disasters, or climate change, often result in rapid and unpredictable population displacement. For organisations like Save the Children, understanding where and when people will move, and in what numbers, is paramount for effective planning and resource allocation. Without accurate predictions, aid efforts can be delayed, misdirected, or insufficient, leading to increased suffering for vulnerable populations. The challenge lies in moving beyond reactive responses to proactive, data-driven strategies.

A Partnership for Impact: Flee, FabFlee, and Save the Children

Recognising the potential of advanced simulation to transform humanitarian planning, a significant partnership was forged between the developers of Flee/FabFlee and Save the Children. Save the Children, a leading independent organisation for children's rights, has extensive experience in responding to humanitarian emergencies globally. Their deep understanding of on-the-ground realities, combined with the computational expertise behind Flee and FabFlee, created a powerful collaborative partnership.

The primary goal of this collaboration was to apply Flee and FabFlee to real-world displacement scenarios, enabling Save the Children to:

• Anticipate future displacement patterns.

• Identify potential bottlenecks and areas of high need.

• Optimise the deployment of resources, including food, shelter, medical supplies, and personnel.

• Develop more resilient and effective humanitarian response plans.

Application in Practice: Simulating Crisis Scenarios

Through the partnership, Flee and FabFlee were deployed to simulate various hypothetical and ongoing crisis scenarios. For instance, in regions experiencing conflict, the models were used to predict how populations might move from affected areas to safer zones, considering factors such as conflict intensity, available routes, existing infrastructure, and the presence of aid.

The FabFlee plugin proved particularly beneficial here. By integrating Flee with FabSim3, the team could:

• Run multiple scenarios rapidly: Explore different "what-if" situations by varying input parameters (e.g., conflict escalation, border closures).

• Handle large datasets: Incorporate vast amounts of geographical, demographic, and conflict data into the simulations.

• Automate complex workflows: Streamline the entire simulation process, from data input to output analysis, reducing manual effort and potential errors.

Furthermore, to enhance the reliability and understanding of the Flee model's predictions, sensitivity analysis was regularly conducted. This process involved systematically varying input parameters to understand their influence on the model's outputs. For this, EasyVVUQ, another component of the SEAVEA toolkit, was utilised to efficiently manage the numerous simulation runs required for comprehensive sensitivity analysis. The QCJ Pilot Job component of the SEAVEA toolkit played a key role in scheduling and executing these extensive computational jobs on High-Performance Computing (HPC) resources, ensuring that the analyses were performed quickly and effectively. This rigorous approach allowed the team to identify the most influential factors driving displacement and quantify the uncertainty in their predictions, thereby improving the robustness of the model's insights.

This allowed Save the Children's analysts to quickly generate and interpret complex displacement forecasts, which were then integrated into their strategic planning.

Conclusion and Future Outlook

The partnership between the developers of Flee/FabFlee and Save the Children, underpinned by the capabilities of the SEAVEA toolkit (specifically its FabFlee, FabSim3, EasyVVUQ, and QCJ Pilot Job components), exemplifies how advanced computational modelling can be successfully applied to real-world humanitarian challenges. By providing predictive capabilities for human displacement, Flee and FabFlee have empowered Save the Children to develop model-informed crisis responses.

This collaboration highlights the importance of interdisciplinary partnerships, combining scientific innovation with humanitarian expertise. As global challenges continue to evolve, the integration of applications like Flee into humanitarian operations, supported by toolkits like SEAVEA, will become increasingly important for building more resilient and effective aid systems worldwide. The SEAVEA toolkit remains committed to fostering such collaborations, pushing the boundaries of what is possible in humanitarian response through cutting-edge technology.