Villoria Hernandez, PabloMariñas Collado, IreneGarcía Sipols, Ana ElizabethSimón de Blas, ClaraRodríguez Sánchez, María Cristina2023-11-212023-11-212023Villoria Hernandez, P., Mariñas-Collado, I., Garcia Sipols, A., Simon de Blas, C., & Rodriguez Sánchez, M. C. (2023). Time series forecasting methods in emergency contexts. Scientific reports, 13(1), 16141.2045-2322https://hdl.handle.net/10115/26237The primary challenges in managing fire emergencies involve identifying fire hotspots, locating the Emergency Intervention (EI) team, monitoring fire progression, and selecting evacuation routes. To address these issues, HelpResponder was developed as a solution capable of detecting focal points in hostile environments affected by high temperatures and low visibility due to fire. The research involves assessing various models to determine the most accurate predictor of measured CO2 levels, utilizing variables such as temperature, humidity, and air quality from sensors in a fire tower. Statistical methods, including ARIMAX, KNN, SVM, and TBATS, are employed to adjust and model these variables. The study introduces a novel enhancement to SVM by incorporating explanatory variables with a temporal structure. Furthermore, the research demonstrates that combining multiple models yields the most effective forecasting results. An additional contribution is the creation of a compact prediction system designed specifically for energy efficiency, conserving battery power. The system has undergone testing and validation in hostile environments, such as buildings, to simulate real emergency scenarios. Its successful validation suggests that it can significantly enhance emergency response efforts, allowing firefighters to act promptly, thereby mitigating risks associated with inadequate information and improving the efficiency of tactical operations. This advancement has the potential to save lives in emergency situations.The key issues in any fire emergency are recognising fire hotspots, locating the emergency intervention team (EI), following the evolution of the fire, and selecting the evacuation path. This leads to the study and development of HelpResponder, a solution capable of detecting the focus of interest in hostile spaces derived from fire due to high temperatures without visibility. A study is conducted to determine which model best predicts measured CO2 levels. The variables used are temperature, humidity, and air quality, obtained from sensors installed in a fire tower. The statistical methods applied, namely ARIMAX, KNN, SVM, and TBATS, allow the adjustment and modelling of the variables. Explanatory variables with temporal structure are incorporated into SVM, a new improvement proposal. Moreover, combining different models showed the best efficiency in forecasting. In fact, another contribution of our work lies in offering a small-scale prediction system that is specifically designed to save batteries. The system has been tested and validated in a hostile environment (building), simulating real emergency situations. The system has been tested and validated in several hostile environments, simulating real emergency situations. It can help firefighters respond faster in an emergency. This reduces the risks associated with the lack of information and improves the time for tactical operations, which could save lives.Atribución 4.0 Internacionalhttp://creativecommons.org/licenses/by/4.0/Statistics, Electrical and electronic engineeringinfo:eu-repo/semantics/article10.1038/s41598-023-42917-1info:eu-repo/semantics/openAccess