Traffic lights are a strange paradox on our road network. It’s hard to imagine the road without them, and at the same time it’s costing a lot of effort to keep them in good shape. When was the last time traffic control lights underwent a major overhaul? It's time we create a new generation traffic lights.
“Slipshod management of traffic lights leads to unnecessary accidents”, a headline in a Dutch newspaper announced some time ago. The occasion for the article was a study, which showed that 50 percent of road managers barely takes a second look at traffic lights as soon as they have been installed. This means that even though the traffic situation may have changed drastically over the course of time, the lights continue to switch from red to green in the same way forever. As a result road users face unnecessary waits at a red light, and this in turn leads to unnecessary accidents, according to traffic experts quoted in the newspaper article. “If you have a feeling you’re waiting for nothing, this diminishes the credibility of a traffic light”, says Eric van Berkum of the TU Twente University. According to experts quoted, the lack of maintenance is due to the high costs involved.
Purpose of traffic control signals
If traffic lights are indeed in such bad condition, shouldn’t we take the opportunity to have a discussion about them and about the way they are used? It starts with the name. Use the word ‘traffic light’ in the presence of a traffic expert, and they will probably correct you: the devices are actually called traffic control signals. In fact they are complex systems including a control computer and its housing, the poles and the lanterns with traffic lights for the red-amber-green signals, the detectors (often loops in the road surface) that feed the control computer with signals about the presence of road users, and all accompanying power supplies, cables, and leads. And before you know it, you’ve stopped talking about the purpose of traffic control signals and are getting bogged down in technology.
The fundamental purpose of traffic control signals is to “realize optimal handling of traffic at conflict points”. At a time in which we’re moving towards network management, the functions required of traffic control signals become even more all-inclusive. Traffic control signals are also being deployed very deliberately to buffer traffic, to ensure that queues downstream don’t become so long that vehicles have to stop on intersections, thus blocking traffic crossing the intersection in other directions (‘blocking back’). Or to enable ramp metering to prevent congestion on the motorway.
Conversely, traffic control signals can also be used to ramp meter the inflow into a city from a motorway. In short, traffic control signals are a powerful control instrument in network management, and this requires that we focus on functionality – not on the technology used. Ultimately, what traffic lights do is to instruct road users to stop and wait for the sake of a higher network goal.
In the past time traffic control signals were control instruments that perform their task in isolation: realizing optimal handling of traffic at conflict points. In some cases, they were linked through coupling cables, so that green waves could be created along the road section covered by the coupling cable. Practically all settings are defined by the technology as it has been deployed along the roadside.
Later the first traffic signal control rooms came up to monitor technical conditions of the traffic control signals. These made it possible to do functional adjustments by reconfiguring traffic control schemes to make them better attuned to new traffic situations. Many road managers now have their own traffic control room, in which a traffic signal control room occupies an important position in addition to a network management system (NMS). The NMS can be used to activate control scenarios whenever the traffic situation changes, which leads – through the traffic signal control room – to the reconfiguration of current control schemes, or the deliberate choice to deploy a different control scheme that is more suited to the traffic situation at the time. This is the path we should be going down, and it is an opportunity for further growth. This means that the functional condition of traffic signals isn’t checked and adjusted every few years, but continually throughout the day and the week.
Cost is, once again, a limiting factor. In order to realize this, traffic control signals must first be taken out of their isolation; they have to start communicating across long distances with the traffic signal control room. This can be done using copper wire or fiberglass, but wireless communication technologies, or a combination of both, are a possibility as well. The UK already has excellent examples of traffic control signals that are part of a mesh network through wireless communication technology. In this case only one traffic light is linked to fiberglass and executes all communication between the mesh network and the control room. In addition, traffic control signals have to be equipped with a whole series of traffic control schemes to be able to meet a varied set of possible traffic situations. Ideally, communication between the control room and the traffic control signals takes place using the open IVERA protocol, with data from the traffic control signals extracted in open VLOG format, which gives the traffic control room an indication of the number of passing road users. Older traffic control signals will have to be adjusted technically to enable this.
System set-up of traffic control signals
Adding this kind of technology to traffic control signals’ existing system set-up, significantly increases their added value (certainly at network level), but it doesn’t really bring down costs. To do that, we have to look at the system set-up itself. When was the last time real innovation of traffic control signals took place? Why can’t traffic control signals functionally maintain themselves? In an era in which we’re working on autonomous vehicles, it’s difficult to retain the existing approach where traffic control schemes are made manually, compiled, and then loaded onto the traffic control signal on the control computer. Every step in this process costs time and money. Moreover it results in traffic control schemes that are out of date and do not grow along with changing traffic situations. In other words: we want traffic control signals that are autonomous in functional maintenance.
We also want traffic control signals whose system set-up is geared to the great developments of our time, such as cloud computing, mesh networking, and the internet of things. Developments that have brought innovation and reduced cost in many fields, but are only trickling down into the world of traffic. And we want traffic control signals that don’t just look like traffic lights to road users, but that also explain why road users have to stop and wait. Socially communicative traffic control signals, that make it easier to use them to buffer the traffic to prevent congestion downstream.
If we are aware that traffic control signals have a particularly useful function in safely and more efficiently using the available road infrastructure, and if we also know that the cost of maintaining this function is too high, then it is our responsibility to engage in innovation instead of cutting off debate by pointing at rules from the past and at road safety. Innovate first; rules can be adjusted afterwards. Road safety is of course a primary concern, but it’s precisely the current situation that is lacking in this respect.