Humanitarian Aid Optimization

Objective : Globally, road traffic accidents cause significant humanitarian, social, and economic costs, resulting in the need to have efficient and fast response mechanisms. Data-based tools can improve humanitarian aid's speed and equity using mathematical modeling, especially optimization, stochastic, fuzzy, and System Dynamics methods. This paper provides a systematic review of the role of these models in helping with post-accident humanitarian strategies and determining key factors that can affect the success of such models due to uncertainty. Methods : A systematic review was performed under PRISMA guidelines using the PICOS framework. Scopus and Web of Science literature were analyzed, focusing on peer-reviewed studies applying mathematical modeling to humanitarian response in road-accident contexts. Models were categorized by data type (stochastic, deterministic, fuzzy), method (exact vs. heuristic), and capability in managing uncertainty and feedback. Special attention was given to System Dynamics, which captures nonlinear feedback loops and time delays in prevention and response systems.  Results : Recent research highlights a shift toward predictive analytics, IoT, and machine learning to improve humanitarian logistics. Stochastic and fuzzy models effectively address real-world uncertainties, while dynamic and feedback-based models, particularly SD, outperform static ones by enhancing resource allocation, reducing response times, and strengthening decision-making. Conclusion : The mathematical modeling (in particular, with integration into the System Dynamics) demonstrates the possibility of humanitarian aid optimization in road accident handling. The paper highlights evidence-based, adaptive, and feedback-driven solutions through real-time information and uncertainty modeling to develop resilient, efficient, and scientific information-informed emergency response systems.