Aim: The aim of this article was to compare the evacuation times obtained from public buildings, using selected mathematical models, with times of evacuations carried out experimentally.

Introduction: In the article, various mathematical models are presented in order to prove their use in fire evacuation time estimates. These include the critical evacuation time model, the Togava model, the Melenik and Booth model, the Galbreath model and the Pauls model. In order to compare the accuracy of the fire evacuation time estimates obtained by means of the above-mentioned methods, which are meticulously described in professional sources, a variety of real-life evacuations have been analysed, including evacuations from the Institute of Industrial Chemistry in Warsaw, the Public Television building in Lodz, the Marshal’s Office in Lodz, and the Local Fire Rescue Unit in Pabianice. The time checks obtained experimentally during the abovementioned fire drills have been set against the estimates obtained through mathematical analysis and the Pathfinder software computer simulation.

Conclusions: Professional literature on the subject-matter provides various mathematical formulas which can be put into use to quickly estimate the movement time of evacuees. However, the simplicity of the formulas and, therefore, the simplicity of both the analysis and results, can often lead to calculation errors, especially when compared with real-life time checks. The discrepancy between the model-based time estimates and the estimates obtained through real-life experimentation can be rooted in the ignorance displayed by mathematicians as to the necessity of incorporating several critical parameters into their models, such as the structure of vertical/horizontal escape routes and the volume of human traffic within them. Different escape routes and traffic levels may result in highly varied movement speeds and can deeply affect the evacuation time estimates. The mathematical models are, for the most part, oblivious of such detailed aspects of evacuation and only take into consideration the general assessments which can be found in professional printed sources.

Relevance in practice: Evacuation experiments which have been carried out in real life have given us the chance to juxtapose the time checks obtained through mathematical simulation with the factual data, which in turn enabled the critical review of the reliability of the models. What is more, the time estimates have been re-processed with the use of Pathfinder software. In conclusion, the comparative analysis has proven that the Pathfinder software, which incorporates the variable-control mathematical model, provides the most accurate and true to life evacuation time estimates.

Keywords: evacuation, human behaviour, speed of movement, experiment

Type of article: short scientific report