How Traffic Lights Work

Every day, billions of people sit at intersections without thinking about what decides when the light changes. But behind that seemingly simple red-yellow-green sequence lies over a century of engineering, some surprisingly old technology, and algorithms that determine the flow of entire cities. The light you see is really just the final act of a system that starts with buried sensors, travels through a specialized computer, and ends with you either going or waiting.

The short answer

Traffic lights work by receiving input from sensors (or fixed timers) that detect vehicles and pedestrians, processing that input through a traffic controller (a specialized computer), and then commanding signal heads to display the appropriate color. These controllers can operate independently at a single intersection, or they can be coordinated through a central system to optimize traffic flow across an entire network.

The full picture

The origins: from gas lanterns to electric brains

The first traffic signal appeared in London in December 1868, long before the car dominated streets. It was a gas-powered lantern operated by police officers who manually changed the signals to control horse-drawn carriage traffic. That system lasted only 23 days before a gas leak exploded and injured the officer operating it.

The modern electric traffic light was born in 1912, when Lester Wire, a policeman in Salt Lake City, created the first electric version. But it was William Potts, a Detroit policeman, who invented the first four-way, three-colored traffic signal in 1920, solving the problem of intersections with more than two approaches. Meanwhile, Garrett A. Morgan secured a U.S. patent in 1923 for his three-position traffic signal, which introduced the concept of the “all-red” clearance phase that gives intersecting vehicles time to clear the intersection before conflicting movements begin.

Pre-timed signals: the predictable approach

The simplest type of traffic signal operates on a fixed schedule, regardless of actual traffic conditions. These are called pre-timed or fixed-time signals, and they follow a predetermined cycle that repeats throughout the day.

In a typical pre-timed setup, the controller (a specialized computer mounted in a gray box beside the intersection) is programmed with a cycle length, say 60 seconds. During that cycle, each movement gets a specific green time: 30 seconds for the main street, 15 seconds for the side street, and so on. The sequence repeats identically until someone changes the timing plan.

This approach works well in areas with highly predictable traffic patterns, such as central business districts where rush hour follows a reliable schedule. Pre-timed signals are also cheaper to install and maintain because they require no sensor infrastructure.

The limitation is inefficiency. A pre-timed signal gives green time even when there’s no traffic on that approach, while vehicles on a busier approach sit waiting. During unusual events (a concert letting out, emergency detours), pre-timed systems can’t adapt.

Actuated signals: when sensors take over

Actuated traffic signals respond to actual demand rather than following a fixed schedule. They use sensors to detect the presence of vehicles (and sometimes pedestrians) and adjust the signal timing accordingly.

The most common sensor is the inductive loop, a wire embedded in the pavement that detects the metal mass of a vehicle. When a car drives over the loop, it changes the loop’s inductance, and the controller registers a “call” for green time. These loops are typically cut into the asphalt in a rectangular pattern at the stop line and sometimes upstream of the intersection.

Other detection technologies include video detection (cameras with computer vision algorithms that identify vehicles), microwave radar, and infrared sensors. Each has trade-offs: video can be thrown off by shadows and rain, while inductive loops are reliable but require pavement cutting to install.

In a fully actuated intersection, the main street gets green by default, but if no vehicles are detected on the main street for a set period, the controller automatically gives green to the side street. If vehicles are continuously detected, the green time extends automatically within limits.

This is why you sometimes sit at a red light with no cross traffic: the sensor may not have detected your vehicle, or the detection zone doesn’t cover your position.

The traffic controller: the brain at the intersection

The traffic controller is the specialized computer that runs the show. It’s a ruggedized device designed to operate outdoors in extreme temperatures, withstand vibrations from passing trucks, and run reliably for decades with minimal maintenance.

Modern controllers are modular and typically contain a central processing unit, a power supply, and input-output cards that connect to the sensors and signal heads. They’re programmed with timing parameters defined in the Manual on Uniform Traffic Control Devices (MUTCD), the U.S. federal standard for traffic control devices.

The controller enforces several critical safety rules. The all-red clearance interval ensures that after one movement gets red, the intersection remains all-red for a few seconds before the conflicting movement turns green. This gives vehicles time to clear the intersection. The yellow change interval warns drivers the signal is about to turn red, giving them time to safely stop.

Controllers also implement phase skipping: if no vehicles are detected on a particular approach, the controller skips that phase entirely and moves to the next demand.

Coordination: making signals work together

An isolated intersection working optimally can still create congestion if it’s poorly timed relative to its neighbors. That’s where coordination comes in.

The simplest coordination uses a master controller that sends timing synchronization pulses to a group of subordinate controllers. This ensures all signals in a corridor are operating on the same cycle length and that the green waves are offset so that a car traveling at the speed limit can hit green at multiple consecutive intersections.

More sophisticated systems use adaptive control, where central computers continuously monitor traffic conditions across a network and adjust signal timing in real time. Systems like SCOOT (Split Cycle Offset Optimization Technique) and SCATS (Sydney Coordinated Adaptive Traffic System) analyze data from sensors throughout a city and optimize timing to minimize overall delay.

In Los Angeles, one of the first major U.S. cities to deploy adaptive signals, the system manages over 4,500 intersections. According to the Los Angeles Department of Transportation, adaptive signal timing has reduced travel times by 12% across major corridors.

Why it matters

Traffic signals are one of the few pieces of infrastructure where small improvements directly translate to tangible outcomes. The Texas Transportation Institute found that poorly timed signals can waste up to 30% of a driver’s time in urban areas, according to their Urban Mobility Study.

But the stakes go beyond convenience. The Federal Highway Administration estimates that over 50% of all crashes occur at intersections, and many of these are influenced by signal timing. Properly timed yellow intervals reduce red-light running. Properly coordinated signals reduce the stop-and-go driving that increases rear-end collision risk.

Understanding that the light isn’t working against you (it’s not personal, and it’s not random) can also reduce frustration. The person waiting at a red light at 2 AM with no cross traffic may be sitting there because the sensor failed to detect their vehicle, not because the system forgot about them.

Common misconceptions

Traffic signals are timed to make you wait deliberately. This is largely false. While some historic cases of “delayed green” existed to discourage running red lights, modern signal timing aims to minimize overall delay. The frustration of waiting often comes from the controller optimizing for the bigger traffic flow, which may not be you.

If there’s no cross traffic, the light should stay green. With pre-timed signals, this is simply false by design. With actuated signals, it depends on whether the sensor detected your vehicle. If you’re too far back or in an undetected zone, you may wait through a full cycle. Modern systems are improving, but detection gaps still exist.

The yellow light duration is the same everywhere. Not even close. The yellow interval is calculated based on approach speed, grade, and driver perception-reaction time. A yellow on a highway approach might be 5 seconds, while on a residential street it could be 3 seconds. Too short and drivers can’t stop safely; too long and they treat the yellow as a suggestion to speed up.