The Twilight Trap

Quantifying the Impact of Speed and Low Light

Mohd Haniff bin Mohd Azman
S4143511

Introduction

Abstract:

This report analyses the Victorian road crash dataset (2012–2024) to examine a compound-risk hypothesis. The findings show that the specific combination of high speed (100 km/h), low light (Twilight), and regional roads creates a “Twilight Trap” which is a high-risk setting where serious-injury crash rates are higher than the urban-daylight baseline. Motorcyclists are identified as a key vulnerable group. The report concludes with targeted recommendations spanning enforcement, infrastructure, and public awareness.

About the Data:

This analysis uses the Victorian road crash dataset (2012-2024). We are focusing on crashes in 60, 80, and 100 km/h zones, specifically comparing crashes in Urban Melbourne with those in Regional Victoria.

Research Question:

Does the specific combination of a 100 km/h speed zone, Twilight conditions, and a Regional road create a “Twilight Trap”, and if so, which road users are most vulnerable?

Method:

To ensure comparable limits and sufficient counts,60, 80 and 100 km/h zones; ‘Twilight’ groups dawn and dusk to reflect low-contrast light.

1) Serious-injury rate by speed

Findings:

  • What it shows: The relationship between the posted speed zone and the rate of serious-injury crashes across Victoria.
  • Result: The 100 km/h zone has the highest serious-injury rate. This finding is based on a large sample of N = 25,724 crashes.
  • Interpretation: This confirms that high-speed environments are the primary setting for the most severe crashes, providing a clear baseline for our investigation.

2) Light × speed (All Victoria)

Findings:

  • What it shows: The serious-injury rate comparing Daylight, Twilight, and Night conditions across the 60, 80, and 100 km/h speed zones.
  • Result: At 100 km/h, Twilight ≈ 41.8% is 3.6 percentage points lower than Daylight ≈ 45.5%; a similar ordering holds at 60 and 80.
  • Interpretation: Twilight is not universally higher than Daylight in the statewide data; the gap widens at higher speed, but Twilight remains below Daylight overall.

3) Urban vs Regional (60 vs 100; Day vs Twilight)

Findings:

  • What it shows: Serious-injury rates by Urban vs Regional at 60 and 100 km/h for Daylight and Twilight.
  • Result: At 100 km/h, Regional-Twilight ≈ 45% vs Urban-Twilight ≈ 34%—a large regional gap. Within Regional 100, Daylight is similar or slightly higher than Twilight.
  • Interpretation: The Regional environment at high speed is the key amplifier; the Twilight effect is stronger in Regional areas even if it’s not above Daylight everywhere.

4) Compound risk vs baseline

Findings:

  • What it shows: A “risk lift” comparison of our most dangerous cell (Regional–Twilight–100) against the safest baseline cell (Urban–Daylight–60).
  • Result: The “Regional-Twilight-100” cell has a serious-injury rate ×1.37 higher than the baseline. This is a rate of 45.3% vs 33% (from N = 1,698 crashes).
  • Interpretation: This quantifies the compound risk and proves that this specific combination of location, light, and speed is the most critical condition to target for intervention.

5) Who is involved (motorcyclists)

Findings:

  • What it shows: The percentage of crashes in our key cells that involved at least one motorcycle.
  • Result: Regional-Twilight-100 ≈ 10.7%, Urban-Daylight-60 ≈ 12.4%; the largest share is Regional-Daylight-60 ≈ 27.2%.
  • Interpretation: Motorcyclist share is not higher in the high-risk Regional-Twilight-100 cell; the compound risk there isn’t driven by a larger motorcycle mix.

Recommendation

Recommendations:

Based on this evidence, interventions should be highly targeted toward the 100 km/h speed environment in Regional Victoria:

  • Targeted Enforcement: Priorities regional 100 km/h corridors, especially night hours (and the twilight shoulder), when severe-injury risk is highest.
  • Infrastructure Upgrades: Upgrade retro-reflective delineation, edge lines, shoulder sealing and signage on regional highways to improve visibility and driver error margin.
  • Public Awareness Campaigns: Run short, location-specific campaigns for regional drivers on speed–visibility risk, with a note for motorcyclists about conspicuous at dusk/night.

Conclusion

This analysis successfully isolated and quantified the compound risk factors driving the most severe crashes in Victoria. By examining the intersection of speed, light, and road type, the study provides a robust, data-driven mandate for targeted interventions.

Regional 100 km/h roads are the key setting for serious-injury crashes. Compared with the Urban–Daylight–60 km/h baseline, the Regional–100 km/h context under reduced light has a 1.37× higher serious-injury rate (~45.3% vs 33%). Statewide, Night is generally highest and Twilight sits below Daylight at 100 km/h, but the regional + high-speed combination is the main amplifier. These results point to clear, evidence-based priorities: targeted enforcement, low-cost visibility/line-marking upgrades, and rider/driver awareness in regional high-speed corridors.

References

  • data.vic.gov.au. (2025). Victorian road crash data [Data set]. Department of Transport and Planning, Victoria. Retrieved November 1, 2025, from Victorian road crash data
  • Plainis, S., et al. (2006). On-road visual performance during dusk and night. Accident Analysis & Prevention, 38(4), 741–748.
  • Tufte, E. R. (2001). The Visual Display of Quantitative Information (2nd ed.). Graphics Press.
  • RMIT University (2025). Data Visualization and Communication (2510). Accessed 10 October 2025. https://rmit.instructure.com/courses/140961
  • Fitzharris, M. P., Corben, B., Lenné, M. G., Liu, S., Peiris, S., Pok Arundell, T., Stephens, A., Bowman, D. M., Morris, A., & Tingvall, C. (2022). Enhanced crash investigation study (ECIS): Understanding contributing factors for serious injury crashes using crash chain analysis (Report No. 345). Monash University Accident Research Centre. MUARC Report 345
  • Haworth, N. (2020). Motorcycle safety [Fact sheet]. Centre for Accident Research and Road Safety–Queensland (CARRS-Q), Queensland University of Technology. CARRS-Q Fact Sheet
  • Meuleners, L., Hendrie, D., & Lee, A. H. (2008). Socio-economic status and the risk of road traffic crashes in Western Australia: A population-based data linkage study. Journal of Safety Research, 39(1), 45–51. https://doi.org10.1016/j.jsr.2007.12.007
  • Owens, D. A., & Leibowitz, H. W. (1977). Nighttime driving accidents and selective visual degradation. American Journal of Optometry & Physiological Optics, 54(8), 518–525. ResearchGate PDF
  • Wells, S., Mullin, B., Norton, R., Langley, J., Connor, J., & Jackson, R. (2004). Motorcycle rider conspicuity and crash-related injury: A case-control study. BMJ, 328(7444), 857. https://doi.org/10.1136/bmj.37984.574757.EE