Comparison of the evacuation software "Simulex" with New Zealand conditions

Abstract

Throughout the world, there is a move towards performance-based building codes and standards. There is therefore an increasing need to be able to predict how building occupants would react in a given set of circumstances, in order to demonstrate compliance with such codes. Computer models are increasingly being used to calculate the time required for occupants to reach safety in a given building design. One of these computer models is the evacuation software SIMULEX. This research undertook a sensitivity analysis of SIMULEX to question the sensitivity of the predicted overall evacuations times to the assumptions needed to run SIMULEX. Research was also carried out to assess whether it was possible to predict an average human response time to evacuation calls by comparing simulated evacuation times to actual evacuation times. This research did show that it was possible to use SIMULEX to predict overall evacuation times, but that care and consideration in the choice of assumptions needed to run SIMULEX must be taken, as each assumption had different degrees of influence on the overall evacuation times. It also demonstrated that the choice of assumptions should follow the guidelines in occupancy rates set out in the New Zealand Building Code and should reflect findings from research into human behaviour during fire evacuations. While SIMULEX can be used to predict overall evacuation times, this research was not able to use it to estimate an average human response time to evacuation calls because of the small sample size of four buildings.