PE 2019

Characterization of Roughness on Urban and Low-Speed Roadways

By

Steven M. Karamihas

University of Michigan Transportation Research Institute

With Mark E. Gilbert, Michelle A. Barnes

PE 2019

NCHRP 10-93COTR:Amir Hanna

PE 2019

Valid measurement of longitudinal profile is at the core of the approach.

• Reproducibility/Time Stability

• Versatility

• Diagnostics

Characterization of the roughness will depend on profile, not the roughness source.

• Vehicle response (e.g., ride) is of primary importance.

• Tools are needed to identify roughness sources.

NCHRP 10-93 Approach

PE 2019

•Relate objective measurement of ride vibration on urban and low-speed roads to roughness.

•Use standard measures of “discomfort” caused by vibration.

•Seek correlation to roughness.

Ride Experiment

PE 2019

• 29 Test sections

• 6 routes

• functional class 3 and 4

• speed limit 30-55 mph

Ride Experiment

Source: NCHRP Rpt. 914

PE 2019

Test Vehicles

Source: NCHRP Rpt. 914

PE 2019

Instrumentation: Driver/Vehicle InterfaceAccelerations

Source: NCHRP Rpt. 914

PE 2019

Instrumentation: Profiler

Source: NCHRP Rpt. 914

PE 2019

Accelerometer Output

Source: NCHRP 10-93

PE 2019

SAE 2834/ISO 2631 Frequency Weighting

Source: NCHRP Rpt. 914

PE 2019

SAE 2834/ISO 2631 Frequency Weighting

Source: NCHRP Rpt. 914

PE 2019

“Rough” Ride Metrics

Source: ISO 2631/SAE J2834

rmsaw =1

Na2w (i)

i=1

N

åé

ëê

ù

ûú

1

2

PV = k2xrmsa

2wx + k2

yrmsa2wy + k2

zrmsa2wz( )

12

OVT = PV 2ff + PV 2

sbk + PV 2sbt( )

12

Root Mean Square Weighted Acceleration:

Point Vibration Total:

Overall Vibration Total:

* Stay tuned for “Transient” metrics.

PE 2019

“Golden Car” Model

Sayers, M.W., “On the Calculation of International Roughness Index from Longitudinal Road Profile.” Transportation Research Record 1501 (1995) pp. 1-12.

PE 2019

Golden Car Frequency Response

© Copyright University of Michigan

PE 2019

Correlation to Discomfort

Source: NCHRP Rpt. 914

Left IRI:

MRI:

PE 2019

Correlation to Discomfort

Source: NCHRP Rpt. 914

GCARS35:

GCARVV:

PE 2019

Left IRI versus Floor/Foot Acceleration

Source: NCHRP Rpt. 914

R2 = 0.799RMS resid. = 0.0064 g

PE 2019

GCARS35 versus Floor/Foot Acceleration

Source: NCHRP 10-93

R2 = 0.866RMS resid. = 0.0052 g

PE 2019

GCARVV versus Floor/Foot Acceleration

Source: NCHRP 10-93

R2 = 0.899RMS resid. = 0.0046 g

PE 2019

•Limited test vehicles.

•Other responses.

•Thresholds.

• Passengers.

• Localized roughness.

Technical Issues

PE 2019

IRI Generality

IRI

Small Sedan

12 5

102 5

1002

0

.5

1

1.5

2

2.5

Gain for Profile Slope

Wavelength (ft)

IRI

Sports Car

12 5

102 5

1002

0

.5

1

1.5

2

Wavelength (ft)

Gain for Profile Slope

IRI

Luxury Car

12 5

102 5

1002

0

.5

1

1.5

2

2.5

3

3.5

Wavelength (ft)

Gain for Profile Slope

IRI

Mid Sized/Large

Sedan

12 5

102 5

1002

0

.5

1

1.5

2

2.5

Wavelength (ft)

Gain for Profile Slope

© Copyright University of Michigan

PE 2019

Other Responses

Source: NCHRP Rpt. 914

PE 2019

Other Locations

zs

zurzuf

θs

© Copyright University of Michigan

PE 2019

Thresholds: Meaning of “inches/mi”

25 35 45 55 65 75

60

80

100

120

140

160

180

200

Choppy

Wavy

Standard

Speed (mi/hr)

GCARS (in/mi)

Karamihas, S.M., “Simulation Speed and Its Implications to the Relevance of the IRI.” American Society for Testing and Materials STP 1555 (2012) pp. 248–266.

PE 2019

0

.01

.02

.03

.04

.05

.06

.07

.08

25 35 45 55 65 75Speed (mi/hr)

Choppy

Wavy

Standard

GC RMS Sprung Mass Accel. (g)

Thresholds: Meaning of “inches/mi”

Karamihas, S.M., “Simulation Speed and Its Implications to the Relevance of the IRI.” American Society for Testing and Materials STP 1555 (2012) pp. 248–266.

PE 2019

Transient Ride Metrics

Source: ISO 2631/SAE J2834

Root Mean Quad Weighted Acceleration:

Maximum Transient Vibration:

Crest Factor:

rmqaw =1

Na4w (i)

i=1

N

åé

ëêê

ù

ûúú

1

4

rmsaw,T ( j) =1

Ma2w (i)

i= j

j+M-1

åé

ë

êê

ù

û

úú

1

2

MTV =max rmsaw,T ( j)( ), j = 1,N -M

CF =max aw( j) , j = 1,N( )

rmsaw

Transient vibration if: rmqaw

rmsaw>1.5,

MTV

rmsaw>1.5, CF > 9

PE 2019

MTV/rmsaw, Mid-Sized Sedan

Source: NCHRP Rpt. 914

PE 2019

MTV versus Peak Localized Roughness

Source: NCHRP Rpt. 914

PE 2019

• IRI correlated to measures of ride discomfort on low-speed and urban roadways, but better correlation is possible.

• A shift toward shorter wavelengths improved correlation.

• Optimizing correlation for limited conditions is not recommended.

• Localized roughness must be considered to quantify functional quality.

Summary

PE 2019

• Can a new scale be accommodated?

• Should we avoid a scale in inches/mi?

• Should we be using a relative or absolute measure of localized roughness?

• How shall we establish new thresholds?

• What is a higher priority, functional status or pavement health?

Discussion Points

PE 2019

The Report…...

Download NCHRP Report 914

Thank you!!!!

http://www.trb.org/Publications/Blurbs/179566.aspx

PE 2019

Built-In Roughness: Hit or Miss Utility Cover

Source: NCHRP Rpt. 914

300

250

200

150

100

50

0

Left Roughness Profile (in/mi)

5736057340573205730057280572605724057220

Distance (ft)

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

-0.6

Left Elevation (in)

Pass 1

Pass 2

Pass 3

Pass 1

Passes 2 and 3

25-ft base length

PE 2019

Built-In Roughness: Compound Event

Source: NCHRP Rpt. 914

PE 2019

Built-In Roughness: Compound Event

Source: NCHRP Rpt. 914

4000

3000

2000

1000

0

Right Roughness Profile (in/mi)

49204900488048604840482048004780

Distance (ft)

-4

-3

-2

-1

0

1

2

3Right Elevation (in)

Sunken Utility Cover

Utility Cover

High-pass filtered (100 ft)

10-ft base length

Crowned Intersection Ped.Crossing

Ped.Crossing