A knowledge-based system to represent spatial reasoning for fire modelling

This paper describes the architecture of the knowledge based system (KBS) c omponent of Smartfire, a fire field modelling tool for use by members of th e fire safety engineering community who are not expert in modelling techniq ues. The KBS captures the qualitative reasoning of an experienced modeller in the assessment of room geometries, so as to set up the important initial parameters of the problem. Fire modelling expertise is an example of geome tric and spatial reasoning, which raises representational problems. The app roach taken in this project is a qualitative representation of geometric ro om information based on Forbus' concept of a metric diagram. This takes the form of a coarse grid, partitioning the domain in each of the three spatia l dimensions. Inference over the representation is performed using a case-b ased reasoning (CBR) component. The CBR component stores example partitions with key set-up parameters; this paper concentrates on the key parameter o f grid cell distribution. (C) 1999 Elsevier Science Ltd. All rights reserve d.