🚦 Introduction

A primary method of reporting road traffic by the Department for Transport (DfT) involves estimating Average Annual Daily Flow (AADF) of traffic on UK roads. Counts are manually performed at pre-determined locations (called 'count points'), periodically. Typically, these counts are collected over a 12-hour period, only on weekdays, between March to October. There may be inconsistencies in which roads are counted, year on year.

Utilising the public dataset under an Open Government License, we explore it to understand and present our insights on traffic trends in Scotland (because we are at the University of Edinburgh, naturally) covering a period from 2000 to 2019.

🚩 Data

The file dft_rawcount_region_id_3.csv contains the data for all manual counts performed on Scottish roads from 2000 to 2019.

In this data, a 'link' is the name used to refer to a particular section of road. A given road (i.e. a unique road_name) can contain multiple links. One "count point" is decided for each link (a location at which the count is made for that link), and given a unique count_point_id.

There are 33 columns in the data; the description of what each column represents is given here for reference. It's probably a good idea to have a look at the data first, and come back to these tables if you're not sure about one of the columns.

Column name	Description
count_point_id	A unique reference number for each link (i.e. section of road)
direction_of_travel	Direction of travel: N, S, E, W, or C for 'Combined' (when the direction can't be easily determined)
year	Counts are shown for each year from 2000 onwards
count_date	The date when the actual count took place
hour	The time when the count took place, where 7 represents between 7am and 8am, and 17 represents between 5pm and 6pm
region_id, region_name	Website region identifier, and name of the region that the count point sits within (in the provided datafile, which is only for Scotland, these will always be 3 and Scotland respectively)
local_authority_id, local_authority_name	Website local authority identifier, and name of the local authority that the count point sits within (for instance, Aberdeenshire)
road_name	Road name (for instance M9 or A698 -- see e.g. the green and purple labels in this map). Note that M.. roads are motorways, A.. roads are major roads, B.. roads are minor roads, and C and U roads are very small roads, which are not numbered (i.e. for example, all C roads have road_name == 'C')
road_type	Whether the road is a Major or Minor road
start_junction_road_name, end_junction_road_name	The road names of the start and end junctions of the link
easting, northing	Easting and northing coordinates of the count point location
latitude, longitude	Latitude and longitude of the count point location
link_length_km, link_length_miles	Total length of the network road link for that CP (in kilometres and miles, resp.)

The remaining columns all contain the observed number of vehicles for different types of vehicle:

Column name	Vehicle description
pedal_cycles	Pedal cycles (e.g. bicycles)
two_wheeled_motor_vehicles	Two-wheeled motor vehicles (e.g. motorbikes)
cars_and_taxis	Cars and taxis
buses_and_coaches	Buses and coaches
lgvs	Light goods vans (e.g. commercial vans)
hgvs_2_rigid_axle, hgvs_3_rigid_axle, hgvs_3_or_4_articulated_axle,hgvs_4_or_more_rigid_axle, hgvs_5_articulated_axle, hgvs_6_articulated_axle	Different kinds/sizes of heavy good vehicles (e.g. trucks, lorries)
all_hgvs	All heavy good vehicles combined
all_motor_vehicles	All motor vehicles combined (so all of the above, except cycles)

📊 Results

Here we give a snippet of one of the visualizations we were able to generate using the DfT data.