## How many levels in a road hierarchy?

Introduction
In this note I will use the terms “road hierarchy (RH)” and “functional classification (FC)” as if they mean the same thing – which essentially they do. The apparent difference is that RH by definition says that some roads are more important than others, whilst in FC people assume that some roads are more important than others. Meanwhile, both RH and FC group (or class) roads in a network in terms of their function. The question this note raises is, ‘OK, but how many different groups or classes or levels of road are there?’

The note briefly describes examples of:
• A many levels road hierarchy
The note concludes with some comments.

• There are only four levels in a road hierarchy (freeway, arterial, collector/distributor, local)
• Road hierarchy can be defined in terms of two topics. One topic is “access to property”, the other is “through traffic, movement and speed”
• So far as RH is concerned only these 4 topics are involved (traffic, movement, speed and access)
• “Through traffic”,  “movement” and “speed” each have the same relevance to RH

The figure is most likely illustrative, and not based on measured values. Having said this however,

• The graph is a continuous curve – it  does not show four discrete steps. The continuous curve implies that there can be any number of levels in a road hierarchy, not just four. Indeed, the accompanying text brings in additional road types which do not appear in the graphic (e.g. motorways, limited access roads, streets) and also introduces sub-classes (major and minor arterials, rural and urban arterials)
• There is no obvious reason why increasing through traffic should be directly related to increasing speed
• There is no quantification of the term “through traffic.” For example, perhaps a definition of through traffic on a motorway would be “any vehicle trip which is longer than (say) 100 km” – which would mean that any vehicle trip which partly uses a motorway and which is only 99 km long should be defined as “access traffic”.

The USA’s Federal Highway Administration (FHWA) has a webpage on functional road classification (see here) from which the above figure is taken.

From the FHWA figure:

• There are only three levels in a functional road classification (arterial, collector,  local)
• Functional road classification can be defined in terms of two topics. One is “land access”, the other is “mobility”
• The graph is a continuous curve – it  does not show 3 discrete steps. This implies that there can be any number of levels in a road hierarchy, and just 3. Indeed, the accompanying text on the web page has a table 3.2 which shows four classes and a table 3.3 which shows eight classes.

From the accompanying text

• There is an indication that “mobility” refers to journey distance (“the traffic (i.e., local or long distance…”) and to “the character of traffic service”
• The text implies mobility can also be equated with speed (“once the functional classification of a particular roadway has been established, so has the allowable range of design speed”) – and see also this FWHA website, which says “Mobility … can incorporate a wide range of elements …. but the most basic is operating speed or trip travel time”.

Example 1

In 1991 the UK’s Overseas Development Association (ODA) published a book on the planning and design of roadworks, called “Towards safer roads in developing countries” (ref. 617). In its section on road hierarchy, the document says:

“Developed countries define road networks as a hierarchy in terms of road types, according to the major functions the road will serve. The main basis for classification is whether the road is to be used primarily for movement or for access. Roads can be categorised according to their function or according to their operational characteristics”.

The document suggests the following 5 levels in a road hierarchy (table 3.01):

• Primary distributors
• District distributors
• Local distributors
• Pedestrian streets

Example 2

Tanzania’s  Road geometric design manual (2011 edition) (ref. 294) was prepared through technical cooperration with the Norwegian Public Roads Administration. Table 2.1 of the manual refers to 5 levels in a functional road classification:

6-level hierarchy

The East African Community has been working on the harmonisation of roadway geometric standards (ref. 589). Page 11 of the document discusses FC and says:

“Geometric  design  standards  depend  on  the  functional  requirements  of  the  road.  However,  the functional classification of the road system does not automatically lead to the selection of a design speed and cross section for a specific link in the network.”

And

“Roads may also be classified based on some other criteria but the classification based on speed and  accessibility  is  the  most  generic  one”. (own italics)

The figure which accompanies this text (see below) describes six levels of functional classification.

Example 1

Buchanan’s “Traffic in towns” (ref.602 para. 108) says:

“Basically, however, there are only two kinds of roads – distributors designed for movement, and access roads to serve the buildings”

Example 2

Writing in 1989 (ref. 358.7), Australia’s Ray Brindle said that:

“Because the movement and access functions are seen as being continuous, the road classes are not clearly defined but merge from one to the other”

He refers to four grounds for questioning the classical model of road hierarchy, including:

“It creates a broad range of roads intermediate between major traffic routes and minor streets on which there is a conflict of access and movement functions, leading to lower than desirable levels of safety and amenity”

Brindle  argues in favour of a ‘separate functions’ or ‘two categories’ road hierarchy, where a road has either an access function or a traffic function (see figure).

Example 1

In 1996, A. Talvitie wrote that “For several reasons, a state or country may wish to create more than the three functional classes.” (ref. 370). He also said that

“…. specialized  functional  classes  may  be  desirable in some states or countries because of unique functions served.  Examples include parkways, truck routes, busways and high occupancy vehicle routes, private (access) roads and routes for non-motorized vehicles”.

He offered a more extensive classification system, with eight classes (although the list leaves out any specialized functional classes):

• Arterials:  Motorways and other divided arterials;
•   Principal arterials
•   Minor arterials
• Collectors:
•   Major collectors
•   Minor collectors
• Locals:
• Other:    Bicycle and pedestrian paths

Example 2

Lancashire County Council (UK)’s 2002 document on a “Functional road hierarchy strategy” (ref. 306) has a road hierarchy which lists 13 categories (~ classes).

Comment

• Roads exist for many more reasons than just access and through traffic.
• None of the examples quantify key terms such as “through traffic” or “access”.
• The topics which RH focus on appear to be, essentially, access and speed.
• The suggested, continuous curves are misleading. In the world of real roads, there is not an infinitely large number of possible values for speed, but only around 14. The reason is that posted speed limits are generally restricted to multiples of 10 km/hr .
• Adopting Ray Brindle’s two-functions curve I would argue that at speeds from about70 km/hr and above, the “access” element is zero. Below 70 km/zhr the speed/access curve should be a stepped curve, not a continuous curve, with the steps at intervals of 10 km/hr.
• The distance element of access (how many junctions or access points per km) should be based on the concept of road network mesh.
• There are so many requests for exception from the FC concept for special case roads that the FC concept itself begins to seem inappropriate. Examples of these special case roads include:
• Mr. Talvitie’s examples (quoted above)
• Streets (see for example ref. 446)
• Quiet lanes (e.g. ref. 306)
• Terminology – different authors writing at different decades use the same term but not necessarily with the same precise meaning. Imprecise use of terminology for a desirably precise technology such as highway engineering is a bit dodgy.
• If there are road classes for motor vehicles, there should also be road classes for other vehicle types such as pedestrians, cyclists, horse-riders etc – and in fact, examples of such NMT (non-motorised transport) hierarchies exist.

It seems to me that an air of respectability surrounds a concept (RH / FC) which – although it might be useful – nevertheless uses vague terminology, is not well-quantified, ignores practicalities from the real world and is described differently by different writers. Perhaps we can still quote Ray Brindle, where he said, writing in 1989, that:

“practitioners have tended to expect theoretical road hierarchies to be irrelevant in practice”

References

294 - “Road geometric design manual, 2011 edition”, Ministry of Works, Dar es Salaam 2012

306 – “Functional road hierarchy strategy”, Lancashire County Council, England; October 2002

358.7 – Brindle, Ray “Living with traffic: (7) road hierarchy and functional classification” (1989) re-published Australia, ARRB special report 53, 1995

370 – Talvitie, A. “Functional classification of roads”, paper presented at the annual meeting of the TRB, Washington DC, USA; January 1996.

400 – World Bank report , “Improving the Management of Secondary and Tertiary Roads in the South East Europe Countries”, Transport Unit, Sustainable Development Department; February 2008

446 – Ribeiro, Paulo “A new perspective on street classification towards sustainability”, Proceedings of the 8th WSEAS conference; 2012

589 – East Africa, Preparation of the East Africa Transport Facilitation Strategy, thematic area 1 – standards and specifications, “chapter 2: harmonisation of roadway geometric standards”

602 – Buchanan, Colin “Traffic in towns”, London; HMSO 1963

617 – “Towards safer roads in developing countries”, England, TRRL/ODA; 1991