Turning over a wobbly table.
I feel that the Artists conference will be looked back upon as a landmark event in the movement known as “new urbanism”. Seldom can I have been so richly entertained but it is the new insights into other professional worlds that I will most treasure. Chief amongst these was Phil Jones’ presentation “Research Challenges”. Phil put the engineering and road safety professions on the table and under the spotlight and after twenty minutes the table was decidedly wobbly. He introduced us to the engineer’s raison de etre. Engineers identify a failure mode. They predict its occurrence with a mathematical model and then apply a factor of safety. Next they choose design values and monitor their introduction. Finally they refine the model after catastrophe. In the case of highways their concerns are with structural failure, traffic congestion or road accidents. But there are other failure modes of the highway too, ones that aren’t considered: the visual, emotional or social impact and its contribution to physical or mental health.
Their design values contain underlying assumptions about the importance of certain geometric parameters: horizontal and vertical radii, superelevation, lane width, stopping sight distance and visibility splays. Many of these assumptions are now open to challenge from professionals and not just those of us who daily struggle to cross busy junctions. Take stopping distances for example. The Design Manual for Roads and Bridges provides values related to the speed of vehicles but with an allowance for the reaction time for the driver. Sensible standards it might be thought until Phil showed us that every country had a different stopping distance.
Crest curves are apparently decided by the case of the dead dog, in other words the ability of a driver to execute an avoidance manoeuvre upon encountering an object. Originally this was set at 4 inches (100mm) but as car heights became lower crest curves needed to be longer until the construction industry argued that this would be too expensive consequently without any reference to actual crash causation the height was arbitrarily raised to 6 inches (150mm).
Ok, but what about the visibility requirement at junctions. The set back for the minor road is required to be 9m desirably, 4.5m in difficult circumstances and 2.4m in exceptionally difficult circumstances. But the Transport Research Laboratory found for urban T junctions there was no correlation between horizontal lines of sight and accidents. In fact the real danger appears to stem from traffic speeding towards the junction precisely because it has a good line of sight.
The subjectivity follows through into the road safety audit too. The audit is based upon experience, design engineers are barred from the process, its non-reproducible, it’s a problem/recommendation model, there is no quantification of risk, it provides no means of assessing innovative designs and no factual basis for striking the right balance. As the figures are challenged it all starts to look like a black art. One for the authors of the forthcoming Streets Manual to sort out – I just hope that they talk to Phil Jones for the real world research and Living Streets for a more human centred approach.