|
|
|
|
|
|
|
|
Chapter
3 Effect of Speed Limit Changes
| 3.1 Introduction |  |
This chapter examines the evidence of the effect of the speed limit revision program on accident rates, vehicle speeds, traffic offences and public perceptions.
The third specific term of reference for the Inquiry requires the Committee to examine :
The history of casualty crashes on roads prior and subsequent to speed limit revision.
The VicRoads submission stated that:
Insufficient time has elapsed to gain a meaningful comparison of casualty crashes on roads 'before' and 'after' the speed limit review. To enable a comparison to be made with some confidence at least 2 years (preferably 3 years) of 'after' data will be required. Therefore analysis of this aspect will be investigated at the end of 1997 or 1998. [1]
In response to a question asked at the public hearing about the monitoring of changes since speed limits were increased, Mr. D. Anderson, General Manager - Road Safety, of VicRoads said:
Yes, we have had from 18 months to 2 years and we have certainly collected information throughout that period. The before information indicates that on all those freeways where speed limits changed less than 100 casualty crashes per annum and 8 fatal crashes per annum occurred. It is normal for there to be plus or minus 10 per cent variation in crashes each year on those roads.
It is difficult to determine the sort of change required to conclude that the change in speed limit was the contributing factor. Given the small numbers of crashes we would need about three years to be able to statistically attribute changes to the speed limit alone. There is a lot of other activity occurring on those roads, such as improvements to road-sides and other speed management initiatives. [2]
Later in the public hearing the Chief Executive Officer of VicRoads, Mr C Jordan said:
It is reasonably obvious that no large changes in casualty rates have resulted from the speed changes. That is why the road safety researchers are very reluctant to come to a conclusion on what is a small amount of data when you are trying to measure possibly some small change in one direction or the other. That is what makes it very difficult.
I would imagine that if there were a dramatic change as a result of a change in speed limit policy it would have been very easy to see and we all would have said it would take a lot less than three or four years data to measure it. So we are clearly not in the area of quantum change. If there is any resultant change in road accident patterns it is a very small one that is impossible to measure. There is also the difficulty caused by other factors that change in the same period of time, which makes it hard to isolate a single factor in a cause and effect study. [3]
Other witnesses had a slightly different view. Mr J. Sanderson of the Royal Automobile Club of Victoria (RACV) said (in relation to assessment of the 110 km/h sections):
It [VicRoads] does not have enough information at present to complete that second phase. You really need two years for a system-wide analysis. That is the figure bandied about and I think two years is about right. If you want to get down to link-by-link data you probably need a great deal more than two years of information because of the limited number of crashes on those links. We cannot expect to see a change over one particular link demonstrated for a number of years, but over a system-wide basis we could get it after a two-year standard. [4]
The Victoria Police submission said that:
Insufficient time has elapsed to evaluate fully the crash frequency trends. [5]
In response to a question at the public hearing Superintendent J. Bodinnar said the Victoria Police do not yet have the data or the information technology tools to allow them to give a quantitative assessment, but that the information should be possible in:
About six to eight months at the outside, hopefully three to four months but I would like to stick to eight months. [6]
Dr B. Fildes of the Monash University Accident Research Centre was asked whether he thought two or three years would be the period before analysis of the accident effects of the 110 km/h limit could be done. He advised:
I would think that would probably be about the minimum, but it depends on the number of accidents essentially. What is working against us right now is the general reduction in the road toll. We found in other studies that we had to extend data collection periods because the road toll is falling. [7]
The evidence shows that there was a range of opinions on how soon a statistically valid 'before and after' accident analysis could be undertaken. The Committee notes an earlier VicRoads investigation of the effect of the 110 km/h speed limit on rural and outer urban freeways from 1987-89 used 28 months of accident data for 250-300 kilometres of road. [8] The current length of 110 km/h freeway is over 400 kilometres and at the time of the hearing 20 months had elapsed since the 110 km/h limit was re-introduced on 28 February 1993. Consequently there would appear to be a similar number of 'kilometre-months' of accident data available to determine the effect on accidents on this group of roads.
Given the policy importance of the 110 km/h limit and the fact that the speed change had been operating the longest on this group of rural freeways it would seem sensible to undertake an independent analysis of these freeways as soon as statistically feasible.
In the meantime mathematical calculations should be undertaken to determine how soon a statistically reliable accident analysis of the effect of the overall speed limit changes can be undertaken.
The Committee is concerned that this work has apparently not been done.
The VicRoads submission said that:
Procedures have been established to measure the changes in vehicle speeds on those roads affected by the speed limit revision. Vehicle speeds have been measured by the RACV on sample groups of roads within each speed zone category up to 100 km/h, and VicRoads has collected speed data on roads speed zoned at 110 km/h. [9]
A brief appendix to the VicRoads submission provides an 'assessment' of the information of the 'before and after' measurements. VicRoads note that the assessment does not make statistical comparisons of the data advising:
...due to insufficient information; it only provides trends. [10]
The VicRoads appendix is reproduced in Appendix F.
Recently VicRoads has completed a further round of speed monitoring surveys. [11] The speeds are similar to those of the other 'after speed change' surveys. [12]
The RACV submission contained a lengthy appendix providing extensive details of their 'before and after' vehicle speed measurements in both directions of travel at 24 sites in the Melbourne metropolitan area. Further information on the RACV work was also given in a recent paper presented at the 17th Australian Road Research Board Conference. [13]
The RACV information, including their detailed interpretation is given in Appendix G. The main conclusions that the RACV submission drew from their speed measurement surveys were:
Neither the Victoria Police nor other organisations had any further information on the impact of the speed limit changes on travel speeds.
One expected outcome of the speed limit revision program would be more consistent and realistic speed limits and hence a greater degree of driver compliance with those limits. The vehicle speed measurements give some indication of this impact. A further quantitative measure is the effect on detected speeding offences.
At the public hearing Superintendent J. Bodinnar said that:
The incidence of speeding across the state generally has reduced from around 6.5 per cent of cars checked being above the enforcement threshold down to a current 3.2 per cent above the enforcement threshold. There has been a general lowering of the offence rate per hour as well. We operate speed cameras for about 4000 hours a month and the offence rate per hour is currently between 9 to 11 offences per camera hour, which is consistent with our expectations of only a small proportion of the community breaking the law. [15]
Asked by the Chairman whether that reflected a decline Superintendent Bodinnar said:
Yes, it does reflect a decline from about 6 per cent of offences 12 months ago, bearing in mind the speed limit changes had been ongoing throughout the state during the past 12 months. It is only in the past month or two that the whole state has had consistent regulations. [16]
The Victoria Police also indicated that each month they identified all speed camera sites where more than 10% of drivers travelled faster than the speed limit plus the 9 km/h enforcement threshold. If a particular site appeared regularly on that list then VicRoads and the RACV was asked to investigate whether the speed limit at that site was appropriate. Superintendent Bodinnar said:
We have had four or five zones changed as a result of this process of more than 10 per cent of cars being detected and referring that back to VicRoads and the RACV. [17]
The evidence referred to above relates to the current 500 000 speed camera offences per annum. However in the case of other speeding offences the evidence is somewhat different. The published traffic offence statistics [18] show:
Table 3: Traffic Offence Statistics (Non Traffic Camera)
Offence No. of Offences No. of Offences Percent Change
1992/93 1993/94
Exceed Speed Zone 40 397 46 545 +15%
Exceed 60 km/h 74 668 64 964 -13%
Exceed 100 km/h 59 171 61 234 +3%
Speeding Truck 2 330 2 944 +26%
TOTAL 176 566 176 687 No Change
Whilst the figures show no overall change in non-camera speeding offences the truck and exceed speed zone (predominantly the 70 to 90 km/h) offences show significant increases. The latest monthly figures are: [19]
Table 4: Traffic Offence Statistics (Non Traffic Camera) (Current Financial Year)
Offence No. of Offences No. of Offences Percent Change
July - Dec 93/94 July - Dec 94/95
Exceed speed Zone 21 195 25 374 +20%
Exceed 60 km/h 30 683 34 378 +12%
Exceed 100 km/h 27 211 34 297 +26%
Speeding truck 1 321 1 294 -2%
Total 80 410 95 343 +19%
Whilst the above increases in speeding offences may be the result of increased enforcement effort or effectiveness they could also indicate that driver compliance with the speed limits has not yet improved to the same extent as measured by the speed camera data.
Subjective evidence from submissions and the public hearing indicated that the speed revision program had been well received by the public.
In 1992 the RACV invited members to complete a survey in the monthly magazine Royalauto about their perceptions of appropriate speed limits. Over 7 000 responses were received - by far the highest number of responses ever received for a Royalauto survey. [20] Although the survey cannot be seen as being truly representative of motorists' attitudes it does give the only available indication of drivers' views. The RACV submitted a copy of the survey as an attachment to its submission. [21]
At the public hearing Mr J. Sanderson of the RACV said:
Our survey of members, which was taken before the changes were implemented, shows that those that were promoted were being positively received. We have not resurveyed our members to ascertain their opinions following the change in speed limits, but we have been monitoring speeds and they have not changed significantly in that time. [22]
There are still some instances where the perceptions of local communities about the appropriateness of a particular speed zone do not coincide with those responsible for the uniform statewide implementation of speed limits. However these instances appear to be very isolated.
Evidence on the effect on accident rates because of the speed limit changes is not yet available. The Committee was advised that such research has not been undertaken because the time period since the speed revision has been completed is too short in which to collect reliable data. The Committee accepts this argument but insists that this study should be commenced as soon as statistically possible.
The Committee also notes that the limited amount of information on actual vehicle speeds appears to show that they have not changed significantly. This finding reinforces the view that most drivers have always driven at a speed which they perceive to be a safe speed for the particular road they are travelling on.
The Committee recommends that:
1. VicRoads undertake research as soon as possible into the effects on accident rates following the revision of speed limits.
Footnotes to Chapter 3
Return to top
|
|
|
|
|
|
|
|