Three talks · 2022

Highway logs, QC dashboards and event polygons

Three sessions at the 2022 GIS-T Symposium on statewide road data work for the Arizona Department of Transportation: republishing the State Highway System logs after an eight-year gap, executive QC dashboards for the road network, and turning table-only road records (linear events) into authoritative polygons.

Organization
Gistic Research, Inc.
Presented
GIS for Transportation Symposium, Charleston, SC
Year
2022
Files
5 downloads
8 yrs
gap before the logs were published again
105
feature classes under one QC dashboard

Results

  • ADOT’s State Highway System logs published again.
  • QC for SDE domain errors plus LRS event and route errors.
  • A repeatable workflow from nonspatial linear events to polygons.
  • Five statewide road dashboards; the pavement one moved from Tableau to ArcGIS Online.
  • Hosted two special interest groups: FME in Transportation and DOT Data Visualization.

Stack

  • Esri Roads & Highways (LRS)
  • FME
  • SQL Server
  • SDE geodatabase
  • ArcGIS Online
  • Google Data Studio

The sessions weren’t recorded, so each one is here with its title slide and the full deck.

The 2021 State Highway System Log in numbers

From ADOT’s 2021 State Highway System Log mileage summaries, as of December 31, 2021: the last edition I helped with. Figures as published.

9,414.81
centerline miles, including ramps and frontage roads
30,557.65
maintained lane miles
22,362.67
travel lane miles
124
routes and route sections
3,697.88
centerline miles on federal, tribal, military and park land
15
counties and 7 districts
DistrictCenterline milesMaintained lane miles
Central1,485.006,418.76
Northcentral1,523.664,637.44
Northeast1,427.643,954.59
Northwest1,369.314,396.03
Southcentral1,506.534,860.37
Southeast1,124.793,328.37
Southwest977.872,962.10
Total9,414.8130,557.66

Supporting R&H data production with executive QC dashboards

With Shaun Perfect of ADOT: one dashboard for every domain, LRS event and LRS route error in ADOT’s Roads and Highways data.

From nonspatial events to polygons

A Gistic Research workflow that builds polygon boundaries from nonspatial linear events.

How the polygon workflow runs

Some road records exist only as rows in a table: a route, a start and end measure, and a value. FME turns them into polygons in six steps:

  1. Convert mileposts to route measures.
  2. Locate each event on its route to make a line.
  3. Turn each line into evenly spaced points that keep the event’s value.
  4. Draw a Voronoi cell around every point.
  5. Dissolve the cells by value into one polygon each.
  6. On each run, compare the new polygons with the published ones, retire the ones that changed and add the new.
Four panels on two made-up routes. 1: three sections exist only as route measures, from milepost 0 to 36. 2: each section becomes evenly spaced points. 3: a Voronoi cell is drawn around every point. 4: cells are dissolved by section into three polygons.
The workflow on two made-up routes split into three sections.

The dashboards behind the talks

The talks came out of the day-to-day work. As project manager I ran the work behind five dashboards on the state highway system: traffic monitoring (HPMS), structures, the State Highway System detail log, Roads and Highways data QC and pavement condition, which we moved from Tableau to ArcGIS Online.

Diagram in three stages. Extract: source systems are extracted into a reporting database. Transform: FME workspaces join and clean the tables, build geometry and stage the result in a file geodatabase. Publish: the staged data updates hosted layers in ArcGIS Online, a web map sets how they look, and the dashboard reads the web map.
How the traffic monitoring dashboard was fed, redrawn with generic names.

For the structures data on the linear referencing system, FME ran every hour and wrote an error report for each run.

GIS-T 2022, Charleston.
GIS-T 2022, Charleston.

Downloads