Application of Cross-Correlation Technique to Evaluate Profile Data

1
Application of Cross-Correlation Technique to
Evaluate Profile Data

• Rohan Perera
• Soil and Materials Engineers, USA
• Senior Project Engineer
• Perera_at_sme-usa.com

2
Co-Authors

• Starr Kohn, Soil and Materials Engineers, USA
• Mark Swanlund, Federal Highway Administration,
  USA
• Robert Orthmeyer, Federal Highway Administration,
  USA

3
Smoothness Specifications

• State highway agencies have smoothness
  specifications for new/rehabilitated pavements.
• Contractor must meet specified smoothness.
• Penalties if smoothness is less than specified,
  bonus if smoothness is better than specified.

4
Smoothness Specifications

• Contractors use their own profilers to collect
  data in many States.
• Profiler is certified by highway agency annually.

5
Equipment for Data Collection
6
Traditional Method of Certifying Profilers

• Layout one or two test sections (150 m).
• Collect data with a reference device.
• Collect repeat runs with profiler.
• Evaluate repeatability of profiler IRI.
• Compare IRI of reference device with profiler.

7
Is Traditional Method Satisfactory?
8
New Method for Certifying Profilers

• AASHTO Provisional Standard PP-49.
• Uses IRI Filtered Profile and Cross-Correlation.

9
Use of Cross-Correlation Technique
Profile 1
IRI Filtered Profile 1
IRI Filtered Profile 2
Profile 2
10
How Do You Do Cross-Correlation?
ProVAL Software, It is free!!
www.roadprofile.com
11
Experiment to Investigate Use of Cross-Correlation
Low-Volume Loop, Mn/Road
12
Devices for Experiment

• ARRB Walking Profiler.
• ICC Normal Tire Surpro ICC.
• ICC Normal Tire Surpro WI.
• ICC Wide Tire Surpro.
• Walking Profiler B.
• Ames Lightweight Profiler.

13
ARRB Walking Profiler
14
ICC - Surpro
15
ICC Wide-Tire Surpro
16
Device B
17
Ames Lightweight Profiler
18
Sensors in Lightweight Profiler
19
Sensor Footprint
20
Test Sections
Asphalt and Concrete Test sections, 150 m
21
IRI Values, Asphalt Section
Device IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km)
  1 2 3 4 5 Avg
ARRB Walking Profiler 1.39 1.37 1.40 - - 1.39
Surpro - Wide Tire 1.40 1.41 1.41 - - 1.41
Surpro - Norma Tire (ICC) 1.43 1.41 1.42 - - 1.42
Surpro - Normal Tire (WI) 1.41 1.43 1.44 - - 1.43
Device B 1.38 1.39 1.38 - - 1.38
Profiler (TriOD) 1.43 1.42 1.42 1.42 1.41 1.42
Profiler (RoLine) 1.42 1.41 1.41 1.41 1.40 1.41
22
IRI Values Concrete Section
Device IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km) IRI (m/km)
  1 2 3 4 5 Avg
ARRB Walking Profiler 1.20 1.19 1.19 - - 1.19
Surpro - Wide Tire 1.21 1.22 1.23 - - 1.22
Surpro - Norma Tire (ICC) 1.23 1.23 1.24 - - 1.23
Surpro - Normal Tire (WI) 1.23 1.22 1.24 - - 1.23
Device B 1.18 1.24 1.21 - - 1.21
Lightweight (TriOD) 1.21 1.22 1.23 1.23 1.24 1.23
Lightweight (RoLine) 1.21 1.21 1.22 1.22 1.22 1.22
23
Repeatability Cross-Correlation, Asphalt
PP-49 Profiler Repeatability CC 92
Reference Device Repeatability CC 98
24
Repeatability Cross-Correlation, Concrete
25
IRI Filtered Profile, Device B
26
IRI Filtered Profile, Surpro WI
27
Reproducibility Cross-Correlation, Asphalt
PP-49 Profiler Reproducibility with Reference gt
90
28
Reproducibility Cross-Correlation, Concrete
29
Conclusions

• Cross-Correlation is superior to traditional
  method of certifying profilers.
• Equipment problems not detected by the
  traditional method detected by cross-correlation.
• Experiment validated procedure in AASHTO PP-49.
• Devices must traverse same path for method to be
  successful.
• ProVAL will shift profiles so start locations are
  the same before doing cross-correlation.