On February 28 the Birmingham Datablog Meetup hosted a visit from Vikki Holland and Debbie Whittingham from the West Midlands Fire Service. They answered questions on how the Service uses data in both fire prevention and monitoring its activities. Here is a summary of the meetup.
The Fire Service are one of the most advanced users of data in the public sector. With much of their work centred on fire prevention, they draw on a range of data sources to identify and prioritise different risks in different communities.
The service accesses shared data from agencies as diverse as the Office for National Statistics, the Highways Agencies, Cadcorp mapping data, and MOSAIC, an Experian database normally used to profile consumers.
When it comes to data collected by the Fire Service itself, some distinctions are worth noting, in particular:
- Calls versus incidents: not every call is an incident as WMFS can get multiple calls for the same incident or may challenge a call and not respond if they are not needed i.e. burnt toast.
- Process versus intent: a false alarm is a process, but the intent behind that can vary, from malicious calls, to good intent, to equipment issues (for example a faulty alarm, or one that is badly positioned and keeps getting knocked).
Specialist vehicles are now sent to automated alerts (AFAs) with a legal expert who can advise on procedure.
Data on fire incidents is recorded in a number of ways: on return from an incident, for example, officers will fill in an incident workbook. How much information they provide depends on the incident – for example a false alarm will require fewer fields to be completed than a house fire.
Data is also recorded on dispatch of an engine, and on arrival (a button inside the engine is pressed when they arrive); the time when the engine is released by fire control is also recorded.
Incident data collected before and after the incident by the Service includes:
- 999 calls (including automated alerts, for instance ADT alarms): this includes the method of call (for example an AFA, or automated fire alarm)
- Fire engine movement (when dispatched, arrived, etc.)
- Type of Incident e.g False Alarm, Primary Fire, Secondary Fire
- Size of fire
- Equipment used
- People involved
- Casualties and fatalities
- Damage (this can be fire damage or smoke damage)
- Location in house
- If the fire alarm was activated
Data is sent to the Department of Communities and Local Government (CLG) automatically.
Risk analysis data
The Fire Service also collects and analyses a range of data around risk. Data is used to identify where to put physical resources & people, and focus prevention work). “We wouldn’t carry out a chip pan fire campaign in Solihull for example [because it is not a major cause of fire in that area],” explains spokesperson Debbie Whittingham.
The Fire Service also holds data on its staffing. This includes:
- Roles undertaken
- Service length
- Demographic data (the average age is 42)
Stories in the data?
The discussion highlighted a number of under-reported areas, such as:
- Lesser known sources of fire such as domestic devices (washing machines, tumble dryers, and hairdryers)
- Causes of fires where there are large demands on resources but little danger to human life, such as industrial recycling
- Recruitment and staffing: the service receives 30 calls per week for jobs, but haven’t hired in 4 years. This also means the service is ageing, and unable to respond to criticisms about being ethnically representative.
- Charges to companies responsible for alarms. For example, if they are responsible for a trapped elevator (shopping centres have a responsibility to maintain these, and get charged if responsible)
- Sources of ignition – what catches fire
- Weather and fire: outdoor arson is down, possibly due to recent wet summers; house fires increase because people have the heating on; road conditions can cause Road Traffic Collisions (RTC).
- Special Service Calls – a.k.a. rescues: people trapped in a car in a flood; people trapped in a lift or building; rings trapped on fingers. Delivering babies is not something that happens.
- Fire Damage Reports (FDR): fires are classified as a primary fire if in a property (buildings and vehicles are both classed as property) and a secondary fire if outdoors (e.g. grass), or in derelict buildings or dumped vehicles. If a fire attracts more than five fire engines it becomes a primary fire regardless.
As for the cat stuck up a tree? It has to be stuck up there for two days to be a call – and only then if requested by the RSPCA…
Pingback: Previously on @bhamdatablog… notes on fire service and probation data | Birmingham Datablog