Road Safety Risk Model for Low Volume Rural Networks

Transcription

1 Road Safety Risk Model for Low Volume Rural Networks Paul Durdin Dale Harris TrafiNZ Conference 2015 Transport Futures: The Changing Face of Transport

2 Safe System Signature Projects

3 Context Eastern Bay of Plenty Signature Project

4 Risk Prediction - EBoP 57.9% of rural crashes occur on curves Low volume road network (1500 km) Traditional risk assessment using crash history unreliable A new methodology is required

5 Austroads Operating Speed Model Uses road geometry to estimate 85 th percentile vehicle operating speeds by modelling: acceleration on long straights deceleration on curves Also identifies curve risk (desirable, undesirable, unacceptable)

6 A GIS Approach Input: high quality road centreline

7 A GIS Approach Divide centreline into 10m sections Calculate rolling average 30m radius Curve Identification 30m 10m Combine 10m segments where: radius <= 500m and curve direction does not change

8 A GIS Approach Curve Identification R=45m Illustration R=28m R=99m R=33m R=41m R=22m R=30m R=24m

9 A GIS approach - Identify Operating Speed Sections

10 A GIS Approach Identify Maximum Speeds Terrain (and grade %) Flat (<2%) Undulating (2-4%) Hilly (5-7%) Mountainous (>=8%) Curvature Straight 110 km/h 110 km/h 95 km/h 90 km/h Curved 110 km/h 100 km/h 95 km/h 90 km/h Winding 90 km/h 90 km/h 85 km/h 80 km/h Tortuous 75 km/h 75 km/h 75 km/h 70 km/h Maximum (desired) speed derived as a function of curvature and terrain

11 A GIS approach calculate operating speeds Speeds modelled in both directions

12 Austroads Curve Context Definitions Within context: No limit (no deceleration) Desirable (some deceleration) Out-of-context: Undesirable Unacceptable Exit speed Approach speed Curve radius

13 Calibration against State Highway Data

14 SignatureNET: High Risk Curve Identification

15 SignatureNET: High Risk Curve Identification

16 SignatureNET: High Risk Curve Identification

17 SignatureNET: High Risk Curve Identification

18 Safety of Out-of-Context Curves 6,985 curves identified 66.6% of loss of control crashes occurred on the 20.3% of curves classified as unacceptable or undesirable

19 SignatureNET web viewer

20 Further Analysis and Validation of the Speed Model Applied to the Top of the South network (Marlborough, Nelson & Tasman) 3,382 km of rural roads compared to ~1,500 km in the Eastern Bay of Plenty 21,158 curves in the Top of the South region 18.9% of curves classified as unacceptable or undesirable 55.8% of loss of control crashes on these curves

21 Applications of the Speed Model Exploring the potential road safety benefits of speed management Intrinsic link between speed and crash outcomes Nilsson s Power Model Informs the speed management response and priority for intervention.

22 Key Points Supports prioritisation of targeted road safety interventions SignatureNET as single source of road safety information for all project partners Demonstrates innovation within a safe systems project

23 Key Points Network screening tool developed specifically for lowvolume rural networks. Identifies curves by targeting to risk = proactive Assists Road Controlling Authorities make informed decisions about road safety.

24 Paul Durdin T M E Paul.durdin@abley.com Contact the office T M +64(0) (0) W Auckland Level 8, 57 Fort Street P O Box Auckland 1142 New Zealand Christchurch Level 1 and 2, 30a Carlyle Street. PO Box 25350, Christchurch 8144 New Zealand