Evacuation of People with Visual Impairments

Fire have always been a threat to human beings and claim lives every year. A lot is done to ensure fire safety in our buildings an structures, but fires still occur and lives are lost. In the past decades there has been a trend towards more and more complex buildings, which challenge fire safety engineers and the prescriptive fire safety codes. Consquently, performance based fire codes is developed and implemented in countires around the world. Performance based codes allow for use of engineering tools and calculations. Meanwhile, accessibility to the building environment have likewise gained an increased focus in the past decades enabling everyone to enter buildings but is not an insurance for egressibility. Representative evacuation data for vulnerable subpopulations, including elderly, children and people with disabilities, are lacking in literature. It is known that the fatality rate caused by fire for this segment of the population is larger than for able-bodied people. It can therefore be questioned whether our buildings provide a sufficient safety level for this group of people. The aim of the PhD Study is to increase knowledge and data on evacuation characteristics of vulnerable people and with a special focus on blind and visually impaired people. An experimental program is designed to obtain data on walking speeds horizontally and descending stairs, interaction between participants and their interaction with the building environment. Experiments are conducted in different buildings including office buildings, institutional building and a tunnel. In total 148 people have participated in the experiments. Parallel to the evacuation experiments participants are interviewed about their experience with the experiments but also thei use of different building types and the difficulties they meet. The interview study revealed that people with disabilities visit all kind of buildings. It is therefore not possible to neglect their presence in buildings while ensuring equal egress for all occupants. It is also found that building elements such as stairs, signage, doors e.t.c., which are essential elements for a safe and easy evacuation, challenge the movement for occupants having an impairment. Quantitative results on reaction times, walking speeds horizontally and descending stairs are also obtained from the experiments. It is found that N&M model correlates with results found able-bodied. This is expected since the model is based on data for this group. Likewise, the theoretical model is conservative estimate for the hearing impaired participants moving on horizontal planes. For all tested subpopulations the walking speed decreased with increasing person density, but the visually impaired participants are least affected and could maintain a higher walking speed for a longer time. There is found a correlation between the reaction time and how close people are seated. A correlation between reaction time and subpopulation could not be documented. Observations made during the experiments shows that participant generally behave altruistic and assisted fellow participants to evacuate. There are examples of elderly people guiding and assisting children to evacuate, deaf people assisting mobility impaired people, and able-bodied individuals who lend a helping hand to a blind participant and guide him to the place of safety. The heterogeneity of the test sample is also investigated. Comparing the total egress time for a heterogenous test population with a homogenous test population only comprising able-bodied adults it is found that it took twice the time to evacuate the heterogenous group. This result clearly demonstrates that the composition of occupants and their respective evacuation characteristics highly influence the evacuation flow. Hence, precautions should be made while selecting the occupant distribution for fire safety calculations.