In my last post I talked about geometric parameters and said that it would help if the writers of standards used a common layout to present the values they suggest for a particular parameter. The post used examples of data on the parameter “longitudinal gradient” (see here). An engineer might think that, once he/she had found a table with values for the parameter they were interested in, that would be it. Unfortunately things aren’t that simple.
There are a number of modifying factors which can affect the value of a parameter such as longitudinal gradient. For example, Austroads (ref. 1888), under the heading “factors affecting design decisions”, has eight detailed pages on the topic.
Table 1 shows a starter list of around 18 of these factors (in fact I included a similar list in an earlier post on maximum crossfall (see here).
I included a list of what might be called “modfying factors” in an earlier post on maximum crossfall (see here). Then I wondered which might be the most important ones, and whether any could be ignored. It seems to me that some factors are more important than others, as I suggest in the next table.
Notes on some of the modifying factors
Different standards suggest different values for a parameter such as gradient. One basic problem is that many countries and regions are covered by more than one applicable design standard. For example the ASEAN highways standard covers countries such as Singapore and Malaysia, which have their own national highway standards. European main roads are covered by a UNESCO agreement, although I cannot see national authorities in Germany or the UK putting aside their own national standards and only paying attention to the UNESCO one. England has national standards issued by the Department for Transport, but there are others, published by bodies such as county councils. A second problem with standards are that they do not all reflect current best practice.
Year thedesign standard was published
Values for at least some geometric parameters will change with each new edition of a design standard (otherwise there would be no point in publishing the new edition).
This is arguably the most important modifying function. It relates directly to other functions such as road class.
Design standards often discuss longitudinal gradient in terms of cars and trucks, for example by suggesting that on steep gradients at some point a climbing lane for trucks might be an idea. The same reasoning should apply to other types of vehicle. For example, I understand the maximum grade for cyclists could be from 2% to 5% (2% if a wide range of riders is to be accomodated / Ref. 917, referring to bicycle paths). Perhaps roads which are cycle routes should have a maximum grade / or be supplemented by climbing lanes for bicyclists.
Some geometric parameters are affec ted by climate. For example, in discussing maximum crossfall (superelevation) one USA source says the maximum values in areas where snow, ice occur should be 8% (Ref.831). The same source says that the maximum longitudinal gradient (grade) could be as much as 14% (urban collector, mountainous terrain, 30 km/hr design speed). I would have thought that the same maximum value (8% where snow and ice occur) would apply to both crossfall and gradient.
Range of values
The idea (of using a range of values for a design parameter) has been expressed most clearly in Austroads discussion of the “extended design domain” (see a previous blog post here). Other design standards imply a range of values when they use terms such as “desirable, preferred, absolute maximum”.
Many standards relate some geometric parameters in terms of different types of terrain – usually (bot not always) level, rolling and mountainous. Besides the problem that these terms are not consistent, the consideration of terrain is a substitute for cost cutting (and so also cost/benefit analysis). In a way it prejudges the issue.
Highway section type
Longitudinal gradients also depend on the type of highway section – different maximum values for gradient may apply if the road section is in a tunnel, or on a bridge. For example (ref. 1917) says that “grades in road tunnels should therefore be limited to 3.5% in general”.
By “design element” I mean features such as ramps, hard shoulders, traffic lanes etc. For example, ramps at grade-separated intersections can have different values for maximum gradient for a particular design speed than for the main carriageway (see e.g. Ref. 831).
Most countries have their own design standards each with their own values for gradient. I don’t believe this is necessary (and nor do other people, judging by the number of multi-country standards that there are around). Of course different countries see different types of vehicles on the road, and this might be an argument in favour of country-based design standards), but
- The same can be said for regions within one country (that is, different types of vehicle in different regions)
- Similar if “unusual” types of vehicles can be found in different countries (there are cycle rickshaws in London as well as in Dakka).
- Some countries anyway use the design standard developed for another country.
The ASEAN standards indicate different values for gradient depending on road class, as do the Ethiopia standards.. Austroads says that gradient will ….(also) be based on the relative importance of the road”, and perhaps this importance is indicated by road class. However I am not sure road class is a valid modifyer, since both the importance of a road and the road class is usually indicated by design speed.
The fact that a design standard may be in a language other than English does not make its recommendations any less valid. In other words, German or Norwegian recommendations for longitudinal gradient may be better than those from the US / AASHTO, but few engineers see this, as fewer speak German or Norwegian than do English.
Different standards have different definitions of terms such as “level terrain”, “rolling terrain” (and some don’t define the terms at all). It is surprising that scientists can agree about the term for an object which cannot be seen (such as the quark) but engineers cannot agree on or use on common terms for things which can be seen, such as road gradient and terrain. One result is that it can be risky to mix and match suggestions from different design standards because the definitions and terminology in the standards may be different.
In the sense that the area the road is in might be urban or rural, built-up or open land. I am not sure that this should affect a geometric parameter. A more direct factor is the choice of design speed – if this is 120 km/hr, then why should gradient be different at this speed for an urban and a rural area.
Many design standards around today are disappointing. Some do not give sources for the design values which they recommend. Others do not give any advice on the limitations of the valuse they suggest (advice in the form of say a warning message along the lines of “these values for longitudinal gradient do not apply to roads in tunnels or unsurfaced roads or roads suitable for use by bicyclists and other non-motorised transport”.
A better approach would be for design standards to discuss geometric parameters separately, using a common,4-part layout in each case. The question of allowing for modifying factors could be covered by a list of notes following the main table of values. The 4 parts would then be:
- Table of values
I’ll give an example of such a layout in a future post. If you are interested in any particular geometric parameter, let me know.
831 – USA, “A policy on the geometric design of highways and streets”, AASHTO; 2011
917 – US, “Los Angeles 2010 bicycle plan, technical design handbook”; 2010
1884 – Multi-country, “ASEAN highway standards”; ASEAN, 1999
1887 – Multi-country, “AGRD part 3: Geometric design”, Austroads, 2010
1917 – Australia, “RPDM Chapter23 – Tunnels”, Queensland department of roads, 2006