FlowsBook a callSign inCreate free accountSign up
Population and demographics

How many residents each police station serves, and who is furthest away

Population data for police station planning answers two questions: residents per station and the areas furthest from one. How to work out both, and where it works.

Part of the guide Location intelligence automation

By the xMap team · · 5 min read

Population data for police station planning is used for two measurements: how many residents each station serves, and which inhabited areas are furthest from any station. The first shows where the load is uneven. The second shows where a new station, a satellite office or a patrol base would shorten the distance for the most people. Both need only three inputs: the stations, a population grid and the road network.

This is written for police planners, city safety teams, security consultants and researchers who compare neighbourhoods. One limit comes first: the resident figures used here exist in about 100 countries, among them France, Canada, India, Thailand, Egypt and Australia. They do not cover the United Kingdom or the United States yet, so teams there need their own census figures for the population step.

Start with a station list you trust

Everything depends on the list of stations, and this is the step people rush. A places database has several police types: Police station, Police department, Neighborhood police station, Traffic police station, Police box and more, alongside police academies and officers' housing. Searching the lot gives a list that mixes public counters with buildings no resident can walk into.

If you are the police service, upload your own list to My data: name, position and what each site does. If you are an outside analyst, Nearby places can find the Police station types in a city, but read the result and remove what is not a public station before measuring anything.

Residents per station

There are two fair ways to share residents out between stations, and they answer slightly different questions.

  • By reach. Travel area draws what can be reached from each station within a drive time or a road distance (up to 15 minutes or 10 km for now, and up to 25 stations a run). Demographics then counts the people living inside each area. Areas overlap, so a resident near two stations is counted for both. This shows what each station could serve.
  • By nearest station. Demographics, set to Every hexagon, returns the population grid itself: cells of about 0.1 square kilometres, each with its residents. Find nearest gives every cell its closest station. Adding up the cells per station shares every resident out exactly once. This shows what each station does serve if people go to the closest one.

Demographics also returns nine age groups, so the same run gives children and people aged 60 and over per station. That matters for community policing plans, school liaison and services aimed at older residents.

The areas furthest from a station

The nearest-station method produces the second measurement for free. Every cell now has a distance to its closest station. Sort by distance, keep the cells with people in them, and the top of the list is the answer. Two refinements make it useful:

  • Weight by people. A cell 9 km out with 30 residents matters less than one 5 km out with 2,000. Multiply distance by residents, or simply total the residents beyond a threshold you choose.
  • Group neighbouring cells. Single cells are noise. A cluster of twenty adjoining cells beyond the threshold is a neighbourhood, and it is the unit a planning paper talks about.

On a map, the grid coloured by distance shows the pattern at once: rings of short distance around each station and the pockets between them.

A made-up example

The figures below are invented for illustration. A city in the south of France has four stations and wants to know whether a fifth is justified.

Made-up stationResidents nearest to itAged 60 and overFurthest inhabited cell
Central38,0009,1002.1 km
Port61,00012,4003.4 km
North97,00015,8007.8 km
East44,0008,2004.0 km

Invented figures. Residents are shared out to the nearest station by straight-line distance.

Two things stand out. North serves more than twice as many residents as Central, and its furthest cell is nearly 8 km away. Say 23,000 of North's residents live more than 5 km from it, most of them in new housing along one valley. That is a concrete case: not that the city needs a fifth station somewhere, but that one valley holds 23,000 people who are further from a station than anyone else in the city.

What population figures cannot tell you

  • Residents are not demand. A business district empties at night and fills by day; a nightlife street does the opposite. The grid counts where people live. For how busy a place gets, Mobility Footfall gives people-hours of presence, which arrive about 10 days late.
  • Distance is not response time. Patrol cars answer calls from wherever they are, not from the station. These measures describe access to a station and the fairness of a layout, not how fast a call is answered.
  • No crime data is involved. Nothing here knows where incidents happen. Combine it with your own records before deciding anything.
  • Travel areas use usual traffic, not the traffic at the moment, and not driving with lights and sirens.

Having an AI employee keep it up to date

New housing shifts the picture every year, so the count is worth repeating. In xMap Flows the work is one flow, an ordinary piece of location intelligence automation:

  1. Schedule: once a quarter.
  2. Area of interest: the city boundary, drawn.
  3. Demographics: Every hexagon, up to 5,000 cells a run. A larger city is split into districts.
  4. Find nearest: your station table, with a limit such as only within 15 km.
  5. Calculated field and Filter: mark the cells beyond your threshold.
  6. Show on a map and Make a report: the grid as a heat map, and a PDF for the planning committee.

What lands in the inbox each quarter is a report with the residents beyond the threshold, the districts they are in, and the map. The template People per police station, block by block is this flow ready to change, and Blocks with no fire station within 5 km is its sibling for fire cover.

Questions

How do you work out residents per police station?

Give every cell of a population grid its nearest station and add the residents up per station. Or draw the area each station reaches by road and count the residents inside it, accepting that areas overlap.

Does this work in the United Kingdom or the United States?

Not for the population step. The resident figures cover about 100 countries, including France, Canada, India, Thailand, Egypt and Australia, but not those two yet. The station list and travel areas still work there.

How do I find the areas furthest from a station?

After each grid cell has its nearest station and distance, keep the cells with residents, sort by distance and total the residents beyond a threshold such as 5 km. Clusters of neighbouring cells are the areas to look at.

Is distance to a station the same as response time?

No. Patrols respond from where they are. Distance describes how accessible a station is to residents and how evenly a network is spread.

Can the data show where crime happens?

No. It holds where people live, by age group and sex, and where stations are. Incident records have to come from your own systems.

Templates that do this

Read next

More on population and demographics · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.