6.3.4 and 6.3.5 with Appendix B.4: where the assessor cannot access an occupied space, the count may be estimated from staff familiar with it or from densities in similar nearby spaces, adding to the potential error for computers; sampled estimates must reach a sampling uncertainty under 10 %. Unoccupied spaces without evidence use the default. The default of one computer per 25 m2 of rated area may replace an actual count, including for restricted spaces or spaces occupied in the period but empty at the count, without adding error, but not for spaces with only occasional use such as low-use training rooms, server rooms, boardrooms, meeting rooms or stores. Reasons and calculations for estimates, and the list of spaces on the default, are kept.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.