Search

This paper introduces a protocol for the verification of multi-physics wildfire evacuation models, including a set of tests used to ensure that the conceptual modelling representation of each modelling layer is accurately implemented, as well as the interactions between different modelling layers and sub-models (wildfire spread, pedestrian movement, traffic evacuation, and trigger buffers). This work presents a total of 24 verification tests, including (1) 4 tests related to pedestrians, (2) 15 tests for traffic evacuation, (3) 5 tests concerning the interaction between different modelling layers, along with 5 tests for wildfire spread and trigger buffers. The evacuation tests are organized in accordance with different core components related to evacuation modelling, namely Population, Pre-evacuation, Movement, Route/destination selection, Flow constraints, Events, Wildfire spread and Trigger buffers. A reporting template has also been developed to facilitate the application of the verification testing protocol. An example application of the testing protocol has been performed using an open wildfire evacuation modelling platform called WUI-NITY and its associated trigger buffer model k-PERIL. The verification testing protocol is deemed to improve the credibility of wildfire evacuation model results and stimulate future modelling efforts in this domain.

The analysis of human behavior in fire is a relatively young field, only existing for a matter of decades. For much of this time it was used to support the related engineering process, rather than as a significant pursuit in its own right—to provide support for the assumptions used by engineers, designers and by regulators. Prior to this point, the engineering process excluded the human response from the assessment process altogether. The field originally developed according to two principle objectives, both of which were tied to the practice of fire safety engineering: the ability to establish the importance of human performance and then the provision of key supporting evidence for engineering practice. In both instances, the development of the field was determined by engineering practice, rather than the generation of a comprehensive theory that helped to explain and predict phenomena. This evolution of the field has led to an incomplete, disorganized and disparate understanding of the subject matter: human performance in fire. The lifeblood of any field of study is data—data that bridges the gap between observation, understanding and application. This article, and the project on which it is based, represents an attempt to strengthen the data collection process, the representation of this data, and the dissemination of this data to interested parties; i.e., to strengthen the study of human performance in fire. This will be achieved through the provision of several tools (to aid the collection and presentation of data), that will be combined together in the form of an online data portal. This will benefit the field, allow the development of more refined and more comprehensive theories, and allow for better informed engineering activities.