GOMACO World Index --- GOMACO World 30.3 - November 2002
A GOMACO World interview about the International Roughness Index (IRI) with Gary Hoffman, chief engineer with the Pennsylvania Department of Transportation
GW: What is the International Roughness Index (IRI)?
Hoffman: The International Roughness Index is a standardized measurement of the response of a standardized vehicle to a road profile and road roughness. It's an internationally accepted scale for ride quality.
GW: Who originated this concept?
The concept actually came out of a NCHRP report that was published in the late 1970s. This report really defines specific parameters for a computer-based road roughness response system. It is important to emphasize that we are not measuring road profile here, we are measuring the response of a standardized vehicle to that road profile.
GW: It's a different approach to smoothness measurements?
One of my fellows who does the pavement analysis for us likens it to air temperature versus wind chill factor. Air temperature is really what the road profile is. The wind chill factor is what you feel and IRI is really what you feel, in this case, in the seat of your pants as a response to the road profile.
GW: How is it measured?
It is essentially not measured, it is calculated. What we measure is the road profile, and we measure the road profile with sensors. We have accelerometers on the vehicle and they determine a level plane. We needed a data point, as the vehicle goes up and down. You need a level plane calculated and the accelerometers on the vehicle to do that. Then we have laser equipment mounted on the bumper of the vehicle in the wheel tracks to measure the road profile in both wheel tracks and between the wheel tracks. We have anywhere from three to five lasers on the bumper and the distance between the bumper and the road surface is continuously measured so we can develop a road profile, at least in each wheel track and in the middle between the wheel tracks. We do our calculations every tenth of a mile, every 528 feet (161 m). That's standardized. Once we have that road profile measured, we feed that information into a computer program which has a simulation of a vehicle. That vehicle, a California quarter car, was patterned after a Pontiac GTL. It's a standardized vehicle that everybody uses and you can go anywhere in the United States or a foreign country and, given the same profile trace, calculate the same response using that computer software.
GW: What makes the IRI different from other smoothness indexes?
It is different from all the others because it's what the people perceive. We have in Pennsylvania, for the last six to eight years, been talking about IRI and what people feel in the seat of their pants. The media started using those terms and the common motorist got to understand it. "Hey, it's what I feel in the seat of the pants. It's how my vehicle responds to that road surface undulation." It didn't make any sense for us, at that point, to continue to talk about it from a pavement project pay status or acceptance status when we're working with contractors who continue to use profile measurements. We did have some problems with projects where we had acceptable profile measurements but the ride was not acceptable to the public. We might have had a little chatter in the road because of texture or small undulations that were acceptable from a profile standpoint, but it still created an annoying response in the vehicle. It was one of the reasons that drove us to go to a vehicle response as opposed to just measuring the profile of the surface of the road.
GW: What advantages does this system have over other smoothness indexes?
I think the biggest advantage is it is easily understood by the customer. It's what you feel in the seat of your pants. Another advantage is it's standardized. It's an international standard that can be reproduced in any state or internationally.
GW: That's something other profile indexes haven't had?
GW: Was Pennsylvania using the zero-blanking band before?
We started using the zero-blanking band before we switched to IRI. We went to zero-blanking band probably about five years ago and that was driven by these same projects. Before the zero-blanking band, we used the two-tenths blanking band which says all bumps under two-tenths of an inch aren't counted in the road profile. They were disregarded. You could have a roadway with a zero profile if you took everything out underneath two-tenths of an inch, and yet the public felt it very objectionable for a new pavement. That's what we're building pavement for, is to satisfy our constituents, our customers.
GW: How long have you been using the IRI?
PennDOT has been measuring with the IRI since the mid-1980s strictly as a network evaluation standpoint. We started using the IRI as our spec about three years ago for concrete pavement and last year was the first year we used it for asphalt pavement acceptance as well.
GW: What numbers constitute good ride using IRI specs?
We're using the federal standard that says anything less than 95 inches (2413 mm) per mile is good. We currently have a research project being conducted by Georgia State University. A researcher is doing blind tests with our customers in six different counties in Pennsylvania, rural, suburban, urban, areas on different networks, on interstate, on traffic routes, and on non-traffic routes. It gets the customers perception on what is good, fair and poor from a ride standpoint. We're presuming that our customers think 95 is good, but we want to find that out and the study will help.
GW: The acceptability numbers could change in the future?
Right. We've learned a few things from that study already. One, customers expect a better ride on a interstate or limited access highway than they do on a normal two-lane traffic route than they do on a rural farm to market route. We also found out that suburban and urban people expect a better ride quality than rural people.
GW: How do these numbers compare with the zero-blanking band for achieving rideability?
We have had incentive pay for both concrete pavers and bituminous pavers before. We're continuing with that incentive pay for going to the IRI. Overall, it looks like we're still paying about the same amount of bonuses. We went back and evaluated bonuses from the previous year and maybe one contractor may have gotten a little more bonus and another may have gotten a little less bonus than they got with the zero-blanking band after they switched to the IRI bonus schedule. Overall we're paying about the same total amount of bonuses.
GW: Is the IRI something contractors should be afraid of?
I don't think so. What the IRI is driving towards is a smoother ride from the public's perception. I think most of the contractors in Pennsylvania, particularly the concrete contractors, realize that's the goal that they're striving for too, a smoother pavement to satisfy the motorists. The IRI gives us a standardized tool to measure that and the good contractors who were giving us a good ride from a profile measurement standpoint should be able to give us a good ride from an IRI standpoint.
GW: Do you use the IRI for all paving projects?
Right now, since we're just starting into this, we're just requiring the IRI specs on expressways. Next year we will be moving towards the traffic routes. We use it in both rural and urban areas with the exception that we don't put it on a project where there's a lot of stop and go paving or where there's a lot of manholes and things like that. It's just impossible, if you don't have that paving train moving continuously, to get as good of quality ride as you do where you have the option to keep the paving train moving. We're selective and take that into consideration when we put the ride spec into a project.
GW: The problem with a lot of these profile indexes is, how do you relate to these measurements five years from now? Do you think you'll have that problem with the IRI since it is an international standard?
No, I think we're going to be sticking with this for a long time. We're not going to be continuing to change it around. I think as time goes on and the industry gets better, we'll try to continue to ratchet up the bar.
GW: If this is based on the smoothness of driving over the road, a Cadillac will drive smoother than a '79 Chevy truck over any road surface. How does the IRI adapt to the different vehicles and their drivers' opinions?
Essentially, the IRI is a standardized calculation of a measurement. We'll come up with the same answer all the time. The computer model looks at the vehicle, the chassis of a vehicle and how stiff the springs are in the vehicle so it's a standardized model. It was modeled after a 1980 Pontiac GTO, but it's all relative. Running over that same road now, in the Cadillac with an IRI of 95, and you run over a road with an IRI of 108 in that same Cadillac, you're going to see a distinct difference. Conversely, if you're running over it with a Chevy pickup truck, you're going to have a different response. It's relative. Obviously, what I feel in the seat of my pants, if I jump from my four-wheel drive Z-71 pickup truck into my wife's Lincoln Towncar, is going to be markedly different over the same road.
GW: How have contractors responded to the IRI so far?
Initially, there was some apprehension, but we worked with industry for three years before we implemented this and gathered and compared data. The industry helped us set the bonus standards. We've been working with industry all along and I think there's been fairly good acceptance of it. The way we got into it with the concrete pavers is, the first year, all we did was compare data. We ran the old profile data and the IRI on the same projects without having an IRI spec. Then we showed the contractors what the difference was. The second year we ran the IRI spec, we gave them a bonus but no penalty. It was a win/win for them because they could get 100 percent pay and bonus, but if they fell below a certain amount we didn't penalize them. Then, in the third year, we actually got into bonus and penalty. We eased our way into it.
GW: Who's responsible for testing the IRI of a new roadway?
Pennsylvania is an end-performance state on our specifications. Ride quality is one of the performance characteristics and we require the contractor to do that testing from a project acceptance standpoint. We want them to have the equipment because they run it on a daily basis and they can tell right away when their operation is getting out of sync and they need to make some adjustments. Even though they own the equipment, it has to be calibrated and their operators have to be certified by us to make sure they're coming up with the standardized measurement.
GW: Any final comments?
Ride quality is our number one performance measure in the department. It's something that our customers continually tell us in surveys that's at the top of their list from an importance standpoint on the products that we provide. We look at it very closely when we measure how well we're doing. We program our projects by ride quality. We look at the roughest areas and try to program them first so it is the premiere measure that we use in the department to judge how well we're doing. I think it's also the same measure that our customers use to judge how well we're doing.