The primary finding of the paper "A waypoint based approach to visibility in performance based fire safety design" is the development and validation of a waypoint-based assessment methodology that replaces traditional local point-sampling with a path-integrated evaluation of visibility. This approach represents a shift toward higher cognitive and photometric fidelity in Performance-Based Design (PBD).
The main findings and technical contributions are synthesized as follows:
Traditional fire models calculate visibility as a local cell-based quantity, which the authors argue is inadequate for non-uniform smoke environments. The paper finds that visibility must be calculated as an integrated value of the extinction coefficient (
The paper identifies that visibility is not merely a function of smoke density but is restricted by geometric factors often ignored in legacy models:
-
View-Angle Sensitivity: The model incorporates Lambertian radiation effects, where the perceivable distance of a sign is reduced by the cosine of the viewing angle (
$cos 𝜃$ ). This findings shows that signs may become invisible at extreme angles even if smoke levels are low. - Obstruction Detection: By employing a Bresenham-based ray-casting algorithm, the methodology identifies "concealed" agent cells blocked by architectural elements, ensuring that visibility is only calculated for unconcealed lines of sight.
A critical finding is that traditional local visibility assessments can produce overly optimistic results. The paper demonstrates that Visibility Maps—Boolean matrices identifying safe versus unsafe cells—reveal hazards and "blind spots" caused by room geometry and sign orientation that point-sampling at the agent's head would entirely miss.
The authors find that current PBD practices rely on subjective local performance criteria (e.g., a fixed 10m threshold) that vary between engineers. The waypoint-based approach allows required visibility to emerge naturally from the geometry (the actual distance to the sign), thereby reducing personal bias and increasing the credibility of the building approval process.
The methodology was implemented in the open-source Python package FDSVismap. This tool enables the generation of:
-
Visibility Maps (
$𝑀_{𝑖,𝑗}$ ): Spatio-temporal matrices indicating where the nearest exit sign is perceivable. - Advanced ASET Maps: Spatial representations indicating the exact time at which the visibility criterion to a waypoint is first violated at any given floor location.
In technical synthesis, the paper concludes that visibility maps provide a more realistic and distinct assessment of egress routes by coupling physical fire phenomena (inhomogeneous smoke) with the actual architectural constraints of the visual environment.