Dale Harris

[vc_row][vc_column width=”1/3″ css=”.vc_custom_1453342300079{padding-top: 15px !important;}”][vc_column_text]

Dale Harris

[/vc_column_text][/vc_column][vc_column width=”2/3″][vc_column_text]

Presentation: Caution curves ahead! Modelling vehicle speed and road curvature with GIS

Dale Harris, Senior Consultant, Interpret Geospatial Solutions

STREAM: PROTECT

[/vc_column_text][/vc_column][/vc_row][vc_row css=”.vc_custom_1453341977865{border-top-width: 1px !important;border-top-color: #28b7ff !important;border-top-style: solid !important;}”][vc_column][vc_column_text]

5 Things You Will Learn

  1. A methodology for identifying curves on a road centreline (start, end and radius)
  2. How speeds can be modelled on a road centreline using spatial analysis
  3. How the using geospatial techniques to identify high risk curves is being used for a range of road safety applications
  4. How the identification of high risk curves is being used for a range of road safety applications
  5.  How combining spatial road safety data into a single web viewer supports multi-agency road safety collaboration.

Target Audience

Geospatial and transport professionals with an interest in road safety, curvature analysis, linear referencing or advanced geoprocessing techniques.

Presentation Overview

On low volume road networks, where crashes tend to be sporadic and unpredictable, traditional GIS-based risk assessment methods that rely on crash history can be unreliable. As part of a NZ Transport Agency project, the ability to assess road risk using GIS tools and road geometry was explored and a new GIS-based risk assessment methodology was developed. This presentation will demonstrate a new methodology for identifying high risk curves using geospatial tools to identify curves, calculate curve radii, and model driver behaviour (acceleration and deceleration) along a road centreline. The results of this analysis displayed a high correlation between high risk curves and the occurrence of loss-of-control crashes. This vehicle speed model also has the potential to revolutionise how operating speeds are mapped on New Zealand roads. To identify curves and calculate curve radii, a high-quality road centreline was analysed using linear referencing tools. Vehicle speeds along the centreline were then calculated in Python based on observed driver behaviour, including acceleration on straights and deceleration through curves. Two outputs were generated: a vehicle operating speed dataset and a high-risk curve dataset. The vehicle speed dataset predicts speed in both directions at 10m intervals. The high-risk curve dataset identifies and classifies curves by risk category (no limit, desirable, undesirable, unacceptable). The methodology extracted and classified almost 7000 curves across 1500km of road network. When analysed against actual crash locations, it was found that 67% of crashes occurred on 20% of curves classified as ‘high risk’ in at least one direction. These results have been shared with the local road controlling authorities and will be used to prioritise road safety improvements. The vehicle speed model is also being used as a high-quality alternative to commercial probe-based (eg GPS) operating speed datasets.

Biography

Dale is a Senior Consultant at Christchurch-based Interpret Geospatial Solutions. She has a background in both practical GIS applications as well as transport policy and research. Her core strengths include advanced geoprocessing workflows and web GIS.

[/vc_column_text][/vc_column][/vc_row]