This document describes a possible way of creating <receptor>.txt files for RISC-kit's Bayesian network based decision support system. The approach can be extended to represent buffers around receptors, see Tutorial: Creating a <receptor>.txt for a latent receptor.
Expected Output
The <receptor>.txt files should be comma-separated files with values (no headers) in three columns. Each row represents a specific location in the hot spot site. The first column contains areaIDs, the second column receptorIDs and the third column gridIDs. Moreover, all columns are sorted such that the values in receptorID are ascending.
Prerequisites
- A shape file of the 'sub-areas' of the hotspot. Ideally sub-areas are created such that hazards within are similar. It is not necessary (or very desirable) to divide the hotspot such that each area contains (almost) only one receptor type.
- A shape file of the units of the receptor of interest. A unit could be a house, but also a road section. Note: units cannot be split across areas, but must be uniquely assigned to a single area.
- A hazard.nc file (can be a dummy simulation, as long as it contains the proper grid information)
...