Norwegian experiments on human behaviour in smoke have been re-analysed with respect to evaluating and quantifying performance parameters adequate for 5re safety design according to modern principles of 5re safety engineering. Two generic performance parameters are quanti5ed; the movement speed and the probability of making the correct decision in a decision element of an escape path.
For low location lighting (LLL) systems designed according to the speci5cations of IMO, movement speed might be rather insensitive of the type of wayguidance system, even though higher luminance of electrically powered (EP) than photoluminescent (PL) systems. The human performance of making the correct decision in a decision element was found to vary between 70% and 80%, and also irrespective of the type of LLL system. Other types of wayguidance systems have performance greater than 90% with respect to making the correct decision, but these do not comply with the existing pre-scriptive code speci5cations.
The probability of way5nding and successful escape depends on the correct decision in each decision element in the escape path and identi5cation of exits. Therefore the probability of successful escape diminishes sharply as a function of both the number of decision elements and the performance of the wayguidance system.
Unfortunately there are still missing performance requirements for both the geometry of escape paths and the visibility and recognition of safety signs.