FlowsBook a callSign inCreate free accountSign up
Population and demographics

Using population data to decide where recycling points go

A recycling bank only works if enough people live within a short walk. How councils and operators use small-area population to find the blocks a new point serves best.

Part of the guide Location intelligence automation

By the xMap team · · 5 min read

Recycling points should go where many people live and the nearest existing point is far away. With population counted on a fine grid, you can work out, block by block, how many residents each existing bank or drop-off centre serves and which blocks are left a long walk from any of them. New points then go to the blocks with the most people and the longest distance, instead of wherever a spare corner of pavement happens to be.

This is for council waste teams, recycling operators, glass and textile bank providers, and retailers who host collection points. It explains what small-area population figures are, where they exist, how to combine them with your own map of points, a made-up example in Lyon, common errors, and an AI employee that redoes the analysis whenever your network changes.

Why population, not just a map of streets

Recycling is a walking trip for most households. People carry bottles, cans or old clothes a few hundred metres, rarely more. So the number that matters is not how many points a district has, but how many people live within a short walk of one. A district with ten banks clustered in its shopping centre can still leave thousands of residents far from any of them, while a district with three well-spread banks serves nearly everyone.

Demographics gives residents per hexagon of about 0.1 km², with men, women and nine age groups. Asked for every hexagon, it returns the grid itself, ready to draw as a heat map weighted by people; asked for totals, it adds the residents within a chosen distance (500 m to 8 km) of each point you give it. That second form is exactly the question here: how many people live within, say, 500 m of each existing recycling point.

Three numbers that place a recycling point

  • Residents within a short walk of each existing point. A bank serving very many people is likely to overflow; one serving few may be in the wrong place.
  • Distance from each block to its nearest point. Find nearest takes your own list of points from an upload or a table and adds, to every hexagon or place, the nearest one and how far away it is in a straight line.
  • Residents in blocks beyond that distance. Add up the people in hexagons where the nearest point is further than your target walk. That is the population a new point could reach.

Age groups help too. A block with many residents over 60 needs points closer to home, because carrying heavy glass is harder. A block with many children may suit a point near a school, where parents pass every day.

A worked example in Lyon (made up)

Say a waste operator runs glass banks across Lyon and the city asks where to add 15 more next year. The operator has its current banks in a spreadsheet. All figures here are invented to show the method.

  1. The bank list goes into My data as a table with each bank's position.
  2. An Area of interest covers the city, and Demographics returns every hexagon with its residents.
  3. Find nearest adds each hexagon's nearest bank and the distance in km.
  4. Filter keeps hexagons where the nearest bank is more than 400 m away.
  5. Sort and limit orders them by residents, most first.
  6. Show on a map draws the result as a heat map, with existing banks shown.

Say the result shows that most of the far-from-a-bank residents live in two hillside neighbourhoods and one new housing area by the river. The operator places five banks in each, picks exact spots on public land near shops or stops, and runs the flow again with the 15 new banks added to the table to confirm the gap closed. The council gets a before and after map, not just a list of addresses.

Mistakes that put banks in the wrong place

  • Counting points per district instead of people per point. Districts are administrative lines; residents do not walk to their district's bank, they walk to the nearest one.
  • Using straight-line distance across barriers. A bank 300 m away across a motorway may be a 15-minute walk. Use a walking area where barriers matter.
  • Ignoring daytime population. Residents are where people sleep. A business district has few residents but office recycling of its own; treat it separately.
  • Treating the grid as exact. Small-area population is modelled, and a single hexagon can be off. Look at groups of hexagons, not one.
  • Placing a bank where there is no room. The data shows where people are. It does not know the pavement width, parking rules or who owns the land.

Running it with an AI employee

Recycling networks change: banks are removed, new housing fills in, a supermarket offers its car park. An AI employee can redo the coverage map every quarter on a Schedule, from the same table of points, so the team always plans from the current network.

It is the same flow as the example, with xMap LLM added before the email to write a short note on the five biggest gaps: where they are, how many residents they hold, and the nearest existing bank. Make a report turns the map and the note into a PDF for the council meeting, and Email sends it to the waste team.

What lands in the inbox each quarter is a one-page report: the share of residents within your target walk of a point, the blocks furthest from one ranked by people, and a map you can put in front of a committee. When you add or move banks in the table, the next run shows whether the gap closed.

Questions

Which countries have the population data for this?

Small-area population is available in some countries, among them France, Canada, India, Thailand, Egypt and Australia. It is not available for the United Kingdom or the United States, and the block says so when an area has no data.

How far should people walk to a recycling point?

There is no single standard; many teams pick a few hundred metres for banks people walk to. Choose a distance that fits your city, then measure every block against it.

Can I use my existing list of recycling points?

Yes. Upload it as a CSV or keep it as a table in My data. Find nearest reads the positions and adds the nearest point and its distance to every hexagon or place.

Does the analysis account for people who drive to recycle?

Not by default: it counts residents within a walk. For drive-to centres, use a larger distance or a drive-time area around each centre instead.

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.