Non-residential fire

Variable Importance

The most important factor for predicting the category of Very Low to Very High risk is the Number of Fulltime Employees (2019).
Other highly ranked variables include factors related to sectors of industry:
- Hospitality Services
- Wholesale and Retail
- Shops and Other Retail Outlets
- Office and Administration
There are plenty of data sources that relate to people and dwellings, but less information was available on other factors on the built environment, businesses, etc.
The model’s predicted risk levels are less confident for non-residential fires compared to dwelling fires.

Four LSOAs were removed from the risk assessment as they were outliers in terms of the number of incidents during the sample.
These were LOSAs which include the following locations where a high number of incidents occurred:
- HMP Ranby
- HMP Lowdham Grange
- HMP Nottingham and Nottingham City Hospital
- Queens Medical Centre Hospitals

Short name	Relative importance	Relationship	Origin
Full-time employees 2019	100.00%	Positive	Census
Part-time employees 2019	89.16%	Positive	Mosaic
I Hospitality Services	45.20%	Positive	Experian
G Wholesale and Retail	41.49%	Positive	Experian
Shops and other Retail Outlets	37.03%	Positive	Experian
Office and Administration	36.01%	Positive	NOMIS
Constructed 1950 - 1975	27.57%	Negative	EPC Data
Streets Length Km	27.37%	Positive	HERE
Indoors Sub-domain Score	25.50%	Positive	IMD
Factories and Manufacturing	23.51%	Positive	Mosaic

Using the top ranked risk variables, the model predicts the risk level in each LSOA.
This map summarises the output in terms of whether each LSOA is most likely to be Very Low to Very High risk.
Most of the highest risk LSOAs are in urban areas.