GIS-ing Up the Roughometer

1
GIS-ing Up the Roughometer
2
PAVER Pavement Management System

• Developed by the Corps of Engineers for
  management of military bases
• System for rating pavements using observable
  surface distresses PCI
• Pavement conditions can be statistically modelled
  to predict future condition.

3
Why Pavement Management
4
The PAVER Database
5
PAVER Sections

• A Section is a street segment with the same
• Width
• Thickness
• Traffic volume
• Materials
• Age
• Maintenance history
• PAVER sections often do not end at mapping nodes
  just like the construction projects that created
  them.

6
Why a Roughometer ?

• Mark Funkhouser (KCMO) article.
• Users dont care about structure, they only care
  about ride (bumps).
• Roughness and condition are often correlated.
• IRI was developed by the World Bank, and is a
  global standard.
• Roughometer is fast and easy (even though data
  transcription may not be).

7
Inertial Sensor
8
The Roughometermobile
9
The Roughometer
10
The GPS
11
The Processing Unit
12
Roughometer Survey
13
4 Key GIS Pieces in 2 Stages

• Stage 1 Map IRI Values
• Linear Referencing (Personal Geodatabase)
• Create Route Event Tables from Roughometer Output
  Files (Python)
• Create Maps (Dynamic Segmentation)
• Stage 2 Data for Pavement Maintenance
• Spatially Relate to PAVER (Geodatabase and an
  Access Database)

14
1. Linear Referencing
In The World According to GIS, what is a
Roughometer Survey?

• One survey may cross multiple street segments
• Each survey contains roughness (IRI) values along
  short linear stretches (e.g., 26.4ft)
• Survey starts when vehicle reaches 15 mph stops
  when vehicle slows to 15 mph

15
1. Linear Referencing

• Good fit for modeling with Linear Referencing -
  Routes and Events
• Simple LR - No hatching, route calibration, or
  worry about route measure anomalies, etc.
• Temporary Solution (GPS in Future)

16
1. Linear Referencing
Problem File Naming Convention Unique IDs

• Route Events are Related by Route ID
• 4 Files for each Roughometer Survey (4316 sets)
• Folder Name ltDategt e.g., 20050620 (June 20th,
  2005)
• File Name svltHoursgt ltMinutesgtltSecondsgt_ltSurvey
  Numbergt e.g., sv144031_51.rst
• Sort of Unique, but very long e.g.,
  2005062014403151
• Too Long for Long had to go Double

17
1. Linear Referencing
Created Routes from JOCO Centerline (Kirby
Guillot)

• RTF File
• Instructions
• Work List (Python)

18
2. Create Line Event Tables from Roughometer
Output Files
Used Python to Parse RST (Roughometer Survey
Text?) Files

• RST File
• Roughness Data Centered Along Routes

19
3. Create Maps Using Dynamic Segmentation

• Just Like Add XY Data
• Used Event Offsets (watch out for little jaggy
  spike thingies!)

20
4. Spatially Related to PAVER
In The World According to GIS, what is PAVER?

• Pavement management features
• Line segments that usually cross intersections
  and likely end part way along a centerline
  segment
• Not a Linear Referencing Product Route or Event

21
4. Spatially Related to PAVER
Desired Output

• Spatially Adjusted Mean, Min, Max, Standard
  Deviation - One to One (1 table)
• Hyperlinks in PAVER database to Roughometer
  Output Files - Many to Many (2 more tables)

22
4. Spatially Related to PAVER
Overlay Route Events Tool?

• NO! Event Layers must be based on the same route
  reference.
• PAVER not based on JOCO Centerline (and not
  exactly the OP Centerline either)!

23
4. Spatially Related to PAVER
Created Point Event Tables

• 1 Point per foot
• Near Analysis - Python (Similar to Matchcover
  command, but without any tolerance)
• Nearest PAVER ID stored with each point
• Summary stats on PAVER ID in points table
• Captured the distance of furthest point
  associated with each PAVER segment
• Error Checking and General Data Quality Indicator

24
4. Spatially Related to PAVER
Problem Intersections

• Process 1 Route at a time
• Select PAVER segments with street name
• Perform Near Analysis
• Append to final table
• Run Summary Stats

• Problem 1 Routes did not have street names
  (Python again!)
• Problem 2 Clean up street name data so datasets
  match

25
5. Bonus
Annoying Things!

• Schema Locks and Feature Layers (Delete Tool does
  more than docs say)