PRIME site characteristics and behaviour change
Where have PRIME measures been trialled?
In Phase 1 (2020 to 2022) PRIME measures were installed at 35 trunk road sites in the West Highlands of Scotland.
Full details for each trial site are provided including information about the locations, bend directions, what3words, and Google Streetview links (see: Appendix B).
Sites were defined using the following characteristics:
- speed – taking the 85th% speed for a site, where this was below 50mph it was classed as a ‘slower’ approach and if it was above 50mph it was class as a ‘faster’ approach
- gradient – this was classed as either ‘downhill’, ‘flat’, or ‘uphill’ towards the bend
- approach – this was based on whether the bend for the PRIME installation was preceded by another bend or a straight section of road
These characteristics provide the basis for defining 12 bend types (i.e. speed = ‘slower’ or ‘faster’ x gradient = ‘downhill’, ‘flat’ or ‘uphill’ x approach = ‘from a bend’ or ‘from a straight’). The site characteristics help refine the trial sites into overall categories that help with identifying suitable locations for installing PRIME measures. The knowledge gained from the research provides insights that could assist with choosing sites based on potential behaviour change results.
While a range of research questions have been addressed in Project PRIME there are specific issues that need to be considered:
- To date the PRIME road markings have been trialled on two-lane trunk roads (divided by a centreline). Within the United Kingdom, 5.5m is the minimum overall road width for roads with centrelines. This document provides two designs for PRIME road markings based on different lane widths up to 3.5m and over 3.5m.
- The focus of Project PRIME has been in supporting motorcyclists before they navigate round a bend. As such, PRIME road markings are not envisaged to be installed on bends.
- Project PRIME has not installed PRIME measures on single-track or one-way roads and given the focus of casualty reduction specifically on rural roads which tend to be two-lane roads for traffic in both directions.
- Project PRIME has not installed PRIME measures on motorways or dual-carriageways and it is not envisaged they would be suitable for these situations. It is possible that the PRIME road markings could conflict with traffic spacing chevrons and with bend radii being much greater the likelihood of motorcyclists losing control on bends is reduced
Details for the characteristics of each trial site is provided including information about the speed limits and 85th%tile speed surveys (see: Appendix C).
Where did PRIME measures work best?
With 47,770 motorcyclists observed, the following statistically significant results were observed for rider behaviour:
- speed reductions at 23 trial sites
- positive changes in lateral position at the final PRIME gateway marking (closest to the bend) at 29 trial sites
- positive changes in lateral position at the apex of the bend at 27 trial sites
- reductions in braking at 18 trial sites
- increased use of PRIME road markings by motorcyclists across 35 of the trial sites
The findings from Project PRIME provide an indication of potential behaviour change effects that might be observed on bends with similar characteristics (Table 1).
| Characteristics (Speed - gradient - approach) | Number of sites | Speed | Position at PRIME | Position at Apex | Braking | Gateway |
|---|---|---|---|---|---|---|
| Slow - downhill - from a bend | 2 | NA | Sig | NA | NA | Trend |
| Slow - downhill - from a straight | 1 | Sig | Trend | NA | NA | Sig |
| *Slow - flat - from a bend | 4 | Sig | Sig | Sig | Sig | Sig |
| *Slow - flat - from a straight | 3 | Sig | Sig | Sig | Sig | Sig |
| Slow - uphill - from a bend | 1 | Sig | NA | Trend | Trend | Trend |
| Slow - uphill - from a straight | 2 | Sig | Trend | Sig | Sig | Sig |
| Fast - downhill - from a bend | 3 | Sig | Sig | Sig | Trend | Sig |
| Fast - downhill - from a straight | 4 | NA | Sig | Sig | Sig | Sig |
| Fast - flat - from a bend | 2 | NA | Sig | Sig | Sig | Sig |
| *Fast - flat - from a straight | 8 | Sig | Sig | Sig | Sig | Sig |
| *Fast - uphill - from a bend | 2 | Sig | Sig | Sig | Sig | Sig |
| Fast - uphill - from a straight | 3 | Sig | Sig | Sig | Trend | Sig |
Table 1 provides an overview of the results for each of the twelve site categories. The following legend is used:
- * – indicates the sites where most behaviour change was observed
- Sig– statistically significant results for behaviour change observed
- Trend – trends in the data observed (i.e. not statistically significant but almost)
- N/A – no behaviour change was observed
These results represent the combined results across all the sites in each category. It is possible that where significant results are shown for measures these were not found across all sites in that category. For individual site results please refer to the research reports (see: Appendix D).
The information above may assist anyone looking to install the PRIME traffic sign and road markings to consider where they might be most beneficial.
The results indicate that PRIME measures delivered the greatest behaviour change at the following sites with statistically significant results across all the measures:
- slow – flat – from a bend
- slow – flat – from a straight
- fast – flat – from a straight
- fast – uphill – from a bend
- slow – uphill – from a straight
At these sites a range of encouraging results were observed:
- slow – uphill – from a straight
- fast – downhill – from a bend
- fast – downhill – from a straight
- fast – flat – from a bend
- fast – uphill – from a straight
The fewest results for behaviour change were observed at the following sites:
- slow – downhill – from a bend
- slow – downhill - from a straight
- slow – uphill – from a bend
Analysis to rank site characteristics according to the proportion of outcomes that reached statistical significance (“Sig”), indicated that:
- Straight alignments consistently rank highest, regardless of whether sites are classed as fast or slow.
- Downhill and flat straights show the strongest statistical signals, while bends and uphill environments tend to dilute significance.
It is important to emphasise that even at sites where less evidence of behaviour change was observed, there were no instances recorded of negative behaviours arising from the installation of PRIME measures. This provides evidence that PRIME measures did not have a detrimental effect on rider behaviour.