The TomTom Traffic Index is based on floating car data (FCD). TomTom collects this data from various sources to create traffic services for our clients and customers. In the Traffic Index, we use a representative sample of this data, spanning 551 billion km, to assess and show how traffic has evolved in cities around the globe throughout 2023.

The travel time in different cities is a result of multiple factors, which can be grouped into a) quasi-static factors (e.g., road infrastructure, such as street categories, capacities, and speed limits) and b) dynamic factors (e.g., traffic congestion and changes in flow). The static factors determine the optimal travel time (as shown on the city pages) in a city (or our free flow products), whereas the dynamic factors provide the traffic flow changes – and the sum of both gives us the travel time.

Consumption and emissions

The quantity of consumed fuel, kWh or emitted CO₂ depends on multiple factors, such as the total traffic volume, the driving speed, driving style, the steadiness of the traffic flow (accelerations and decelerations) and the regional vehicle fleet composition (share of different drive and vehicle types, like petrol, diesel, hybrid or EV, and motorcycles, cars, vans, buses and trucks, respectively).

We show what an average vehicle consumes, emits or costs if you were to drive it in any city around the globe. To compare the traffic efficiency with respect to consumption and emission across different cities worldwide, we used the same class of vehicle (BEV, Diesel, and petrol combustion) across all cities (see details below).

The presented differences in consumption and emission of traffic originate from the traffic flow in the given city and not from varying fleet compositions.

What is the methodology to estimate consumption?

TomTom traffic data provides detailed information of traffic patterns (speeds, accelerations, travel times and km driven) with a 1Hz resolution and a share of up to 40% of traffic in city centers.

We combine our traffic data with our map data (slopes, road classes) and the emission and consumption model PHEM^* to estimate consumptions and emissions per km.

Using PHEM, we estimate the consumption for more than 700 million km of floating car data (FCD) traces for petrol, diesel and battery electric vehicles. This data set was the basis for our ML model, which was trained and used to estimate consumption and emissions globally and on a total of 551 billion km of FCD data.

^*Passenger car and Heavy-duty Emission Model – a simulation tool developed by the Graz University of Technology (TU Graz).

Vehicle models

To compare traffic states' efficiency worldwide, we use average vehicles with different drive types, petrol and diesel, and EV.

The vehicle classes in PHEM represent the average vehicle on the European market. We use the EU6ab (petrol and diesel), and the BEV Gen 1 EV. These are comparable to average vehicles from each category.

Fuel prices

We collect fuel prices (diesel and petrol) and charging prices from filling stations and chargers from 55 countries for 2022 and 2023. The prices are then aggregated to a daily country-wide value.

Charging prices

We collect charging prices from chargers across the US and Europe, and aggregate the prices for each country and day. The data is compared and complemented with EV charging prices from publicly available EV pricing literature (Lanz et al and eco-movement).

Note that there are other charging options than those considered in our analysis. For the comparison with internal combustion engine vehicles (ICE), we only take into account publicly accessible regular (AC) and fast (DC) charging into account. We do not take into account any home or subsidized charging.