Let’s just say it right from the start – loops are an incredibly accurate and efficient way to detect vehicles. For decades now, loops have been the standard to which all other detection has been compared. Traffic specifications are written around them, because well-tuned loops provide the high level of accuracy required by almost every facet of traffic management and planning.
But loops are far from perfect. When it comes right down to it, loops aren’t always as accurate as everyone thinks, especially when they are poorly tuned or trying to detect vehicles that change lanes. Buried in the road, loops are subjected to a harsh environment that can be exasperated by freeze/thaw cycles, road surface damage or even changes to a roadway’s configuration. They are difficult to install and maintain, and the process requires lane closures that disrupt traffic and put road workers in danger. They have a high failure rate, and maintaining or replacing loops can be costly.
Despite these serious (and potentially expensive) flaws, there are many in the traffic community who still see loops as the best option for accurate vehicle detection. For some engineers, it’s a simple matter of trust – in their experience nothing matches the stable, predictable performance of loops. For others, it’s a case of “fool me once, shame on you; fool me twice…” as early non-intrusive technologies proved to be disappointingly inadequate loop replacements.
Fortunately, non-intrusive technology has come a long way since then, and many transportation agencies are reevaluating their relationship with loops. The road to finding a viable alternative to loops is not an easy one, however, and traffic engineers are prudent to make new technologies prove their performance before investing in less-than- adequate results. As non-intrusive technologies continue to evolve, could now be the time for agencies to bury loop detectors once and for all?
Loops have been used in the traffic management sector for so long they are considered the safe and familiar choice. Introduced in the early 1960s, embedded inductive loops have been used to detect vehicles in nearly every country of the world. Today, the US Department of Transportation reports that loop detection controls over 73 percent of the domestic market for arterial management, and they are used in many other applications as well – loops are still, by far, the most common method of vehicle detection used for permanent count stations and intersection control.
In simple terms, an inductive loop is multiple turns of insulated wire wound into a shallow slot that is saw-cut into the pavement. This energized coil of wire generates an electromagnetic field, and as vehicles pass through this field, they cause a change in the field’s inductance. Only metal masses above a certain size are capable of triggering this change; loops do not detect vehicle weight, but instead measure the amount of disturbance a vehicle causes in the electromagnetic field.
An electronic device connected to the loop, called a loop detector amplifier, measures the change in inductance, and once the change is recognized, it outputs a signal to a nearby traffic controller cabinet indicating that a car is present. These outputs provide a range of data that includes vehicle counts, volume, headway and gap; when pairs of loops are installed in close proximity, they can also provide accurate per vehicle speeds.
Federal Highway Administration Training Course on traffic detectors — 1993. This training course provides an understanding of how detectors work and the best methods to install, maintain, and repair them.
When installed and calibrated correctly, loops can be quite accurate, but that accuracy is not 100 percent. A 2003 study conducted by the University of Utah’s Traffic Lab found that saw-cut loops had vehicle count error rates of two to three percent on freeways and as much as nine percent at intersections; the error rates for speeds on freeways ranged from 1.2 to 10 percent (“Detection Technology Evaluation,” Martin, Dr. Peter, Associate Professor, University of Utah, 2003).
A 2004 Caltrans report found that lane-changing vehicles could significantly reduce the accuracy of loops: “Our testing has also shown that in highway applications, vehicles changing lanes can cause Inductive Loop Detectors (IDL) to ‘over-count’ by 15 [percent] in the merge lane and an average of [plus two percent], when all lanes are combined, for the full roadway” (Detector Evaluation and Testing Team: “Microwave Vehicle Detection Final Report,” Wald, W.M., HG Traffic Operations Office of ITS Development and Support, Caltrans, January 15, 2004).
The reality is, loops have a lot of opportunities to fail, and frequent maintenance or replacements are often necessary. The two most common reasons why loops fail? Incorrect installation and road deterioration.
Under ideal conditions, loops installed properly in concrete could conceivably have an unlimited life span, as long as there are no disruptions in the pavement. Unfortunately, that is almost never possible. Road pavement typically needs to be milled about every seven years, and loops are often destroyed in the process. Road changes, traffic wear-and-tear and other factors, such as freeze/thaw cycles, are nearly unavoidable, so many in the industry generally accept that the average lifespan of a loop will be approximately five years.
“Loop installation and maintenance is a dirty job, and I've found that younger engineers and especially technicians would love to go with anything else, even if it isn't always as reliable,” says Dan Vincent, a regional sales manager at Wavetronix. Vincent began working in the industry 28 years ago as a loop cutter and has seen firsthand the challenges that loops can create.
“Lack of reliability in loop performance over time is a major concern for planning departments, since precise data is a basic requirement imposed upon them,” Vincent says. “Replacing loops can be a costly endeavor, and this presents a significant maintenance problem for planning departments since many do not have sufficient manpower or funding for proper maintenance.”
Download the entire Traffic Detector Selection Procedure report online
Price is also a big determining factor when it comes to selecting detection for ITS. It is undeniable that loops have the cheapest upfront costs per unit when compared to other traffic detectors. “Bubba” Locke Bowden, a traffic engineer in Montgomery, Alabama, says loops are far from falling out of favor because they are still the least expensive way to get accurate traffic information.
With 30 years in the industry, Bowden is well versed in using loops for presence detection, advanced detection and counts. He has also worked with video, radar and wireless pucks. He agrees that the upfront costs for loops are much less than for other forms of detection, which makes them an appealing choice for budget-conscious agencies.
Long-term costs of ownership are a different matter. When maintenance and replacement costs are factored in, together with the costs of deploying workers, closing lanes, cutting and trenching, the overall costs for loops quickly add up. Trenching alone can be quite expensive, and the costs increase if digging through sidewalks and cement is required. Finally, the total number of devices needed for a particular project should be considered in order to compare the true costs of detection on a per project basis.
Bowden believes loops won’t lose much market share to other technologies until costs become more comparable, but there are some in ITS who admit that loops appear to be cheaper until all of these factors are considered. The problem is compounded when multiple lanes need to be detected; loops, which can only detect traffic at a single, fixed point, require at least a device per lane, and this, too, increases costs.
“I do think that as non-intrusive detection technologies become less expensive, loops will become less commonplace,” Bowden says.
In contrast to loops, non-intrusive detection, such as video or radar, is installed above the ground, usually off to the side of the road or above the flow of traffic. Installing non-intrusive detection can often be done without closing lanes or impacting traffic flow, and any required maintenance can be done more safely and efficiently than with loops. Should a non-intrusive unit fail, it can be quickly replaced; if the roadway changes, non-intrusive units can be quickly reconfigured.
Non-intrusive detection first emerged as a potential alternative in the early to mid-1990s, when a renaissance of sorts occurred in the ITS industry. A few rudimentary radar products entered the market, followed by video detection, which claimed to be the “end-all” to loops. A big rush to move from loops to non-intrusive detection meant that these early radar and video products were not thoroughly tested in all environments. As a result, many early adopters became “beta testers” and were badly burned by technologies that over-promised and under-delivered.
The experience left more than a few traffic engineers with a bad taste in their mouth about adopting new technology, and that hesitation continues to drive much of the ultra-conservative traffic sector. Dan Vincent admits he personally hates loops, but he understands why they are still the standard.
“Many people have been in this industry for a long time, and there is a stigma that comes along with new technology,” Vincent says. “Loops are familiar and moving to a new technology can seem too risky to anyone who already has many years of experience working with loops.”
Still, the benefits of non-intrusive detection are hard to ignore. In addition to installation and maintenance considerations, most nonintrusive detection can monitor multiple lanes of traffic with a single unit. And while performance and accuracy continue to plague many non-intrusive detectors, there are some that are giving loops a run for their money. For example, the Wavetronix SmartSensor HD has consistently proven that its accuracies meet, or even exceed those of loops, even in difficult detection conditions. SmartSensor HD has been tested extensively at installation locations around the world, and has been found to be a viable alternative to loops (see the feature stories).
In the end, familiarity and accuracy may not be enough to justify dealing with the headaches of loops when there are non-intrusive options available that can match their performance. As non-intrusive detection continues to evolve, more and more people are becoming aware of the long-term costs of loops, and planners, engineers and technicians around the world are beginning to rely even more on nonintrusive technology. It’s possible that loop detection will eventually be buried in the past, but a shift like that will take open-mindedness and a willingness to prove the performance of alternative technologies.
It is undeniable that loops have the cheapest upfront costs per unit...[however] when maintenance and replacement costs are factored in...the overall costs for loops quickly add up.
