The idea
Waste collection is often scheduled by the calendar: a truck visits a route on set days whether the bins are full or not. Some bins overflow before the visit, while others are barely used. Fill-level sensors try to replace the calendar with information. If the collection team knows which bins are close to full, it can visit those first and skip those that are nearly empty.
How a fill-level sensor works
Most devices are a small box fixed under the lid or inside the top of a container. The common sensing methods are:
- Ultrasonic ranging. The sensor sends a sound pulse down and measures how long the echo takes to return from the top of the waste. The distance to the waste surface indicates how much room is left.
- Infrared or time-of-flight ranging. The same idea using light, which can work in a narrow beam.
- Weight sensing, used on some containers, but harder to fit to existing bins.
- Tilt and temperature readings, added by some devices to flag a bin that has been knocked over or may be on fire.
The sensor measures a few times a day, since the contents of a bin change slowly. It then sends a short message with the reading. Because bins are everywhere and rarely have mains power, the box uses a battery and a low-power radio, in line with the usual rules of duty cycling to keep battery life long. Low-power networks or cellular IoT links are common for the reporting.
From readings to routes
The simplest use is a threshold. When a bin passes a set level, such as three-quarters full, it is added to the next collection list. More advanced systems build the route from the list, taking into account where the vehicle starts, its capacity, traffic and the order that minimizes distance. Some use machine learning to predict when each bin will fill, based on the history of its readings, so the crew can plan a day or two ahead.
The aim is fewer wasted stops, less fuel, less time on the road and fewer overflowing bins. Municipalities sometimes also use the data to review where bins are located: a bin that fills every day might need to be larger or emptied more often, while one that never fills could be moved.
Where the approach falls short
This is a case where the technology is sensible but results depend heavily on context, so it is worth being direct about limits.
Sensors read the surface, not the contents. A bin with a pile of waste under the sensor can look full while still having space elsewhere. A lid that is propped open, a bag that blocks the beam or a wet or fluffy surface can cause wrong readings. Loose waste bounces readings around.
Collection routes are not always flexible. Trucks may serve a mix of sensor-equipped and ordinary bins. Staff schedules, labor agreements, fixed pickup commitments and legal or health requirements may limit how far routes can change. If a truck must visit a street anyway, skipping one bin saves little.
Savings depend on the starting point. Where bins fill unevenly and routes are long, the benefit may be meaningful. Where collection is dense and predictable, a calendar schedule can already work well. Claims from one city do not automatically carry over to another.
Maintenance is not free. Sensors get dirty, are damaged by compaction or rough handling, and have batteries to replace. A fleet of devices requires device management, including tracking which have gone quiet. Poor data quality, such as a sensor stuck at “full”, can send crews to bins that do not need emptying.
Cost and lifespan. Each sensor adds cost, and the total cost over several years, including installation, connectivity and replacement, needs comparing to the expected saving. The electronic waste generated by retired devices is a consideration as well.
Related approaches
Some cities use sensors on compacting bins, which squeeze waste and so signal fullness differently. Others put sensors on large containers for commercial or recycling collection, where each visit is expensive and the case is often stronger than for small street bins. A cheaper alternative in some places is simply to review pickup frequency and bin sizes using the drivers’ own records.
Vehicle-mounted systems can also help. Trucks fitted with location tracking show the routes actually driven, which is useful for fleet tracking and for checking that the planned route matches what happened.
Practical advice for a trial
A sensible first step is a pilot on a limited set of bins that are known to be a problem, such as popular park or market bins that overflow. Compare overflow reports, number of stops and fuel use before and after. If the numbers improve, expand; if they do not, the issue may be bin size or collection frequency, not information. For residents, the useful signal is still the simple one: report overflowing bins to the city through its normal channels, because sensors may not be installed everywhere.