You are here

Digital cycling planning tools

Providing a realistic picture of bicycle traffic

Digital cycling planning tools

The Bike Citizens Heat Map of Cologne
The Bike Citizens Heat Map of Cologne ©
Digitalization comes with new opportunities for cycling planning. Special traffic models and extensive data collections make it possible to better address the requirements of cyclists.


Nearly all cycling planners have to cope with insufficient data on local bicycle traffic. Stationary permanent count sites are expensive and only map particular network spots; manual counts require much effort, are costly and frequently map a very restricted period of time. Most traffic models are inadequate for cycling calculations and planning, too.

Digitalization now paves the way for obtaining more exact and reliable data on cycling and evaluating these data with the help of specific traffic models. What data are relevant and how to collect and use them will be illustrated in this article. It will also give an overview of other useful cycling planning tools.

Traffic models

Non-motorized traffic has been insufficiently taken into account or even been completely ignored by standard regional transport planning models. One reason is that the existing models were primarily designed for the requirements and calculations of motorized traffic; another reason is that there are just no data available on non-motorized traffic. Local authorities often only have access to the data of a small number of count sites or have to draw on occasional traffic counts. But these data are insufficient for the development of concrete models and forecasts for bicycle and pedestrian traffic.

The requirements of bicycle traffic (and of pedestrian traffic too; but this article focuses primarily on cycling) significantly differ from the requirements of car traffic. Alongside infrastructure elements, minor roads are of interest for cyclists, too, as there is little or no time loss for cyclists using these roads, in contrast to motorized transport users. Shortcuts along sections closed to motorized traffic as well as topography also play an important role for bicycle traffic. Therefore, it is not suitable to use traditional traffic models for cycling planning.

Oregon Metro

Within the scope of a study, the routes of test persons that regularly travel by bicycle in the city of Portland, USA, were tracked with the help of GPS. The collected data made it possible to derive factors that influence cyclists in choosing a certain route and to find out what criteria are particularly important if they are to be willing to accept a more circuitous route. The lessons learned were implemented into the regional traffic forecast model of the local authority for Portland and the region (Oregon Metro). As a result, future traffic forecasts will be able to deliver better estimates on traffic in general and bicycle traffic in particular. It is therefore possible to make more exact forecasts on the benefit and usefulness of investments in different cycling infrastructure elements [Friderich et al. 2011].

Information box: the most important findings of the study by Friderich et al.

In the Portland study of 2007, the routes of 164 cyclists were tracked by GPS for several days. In addition, the test persons were interviewed on socio-economic data and behavioural parameters.

The participants cycled an average of 10 kilometres per day; the average bicycle trip distance was around 4.5 kilometres. Only 5% of the trips were made for purely sporting reasons; the most frequent purpose apart from riding back home was riding to work (25%) [Friderich et al. 2011]. This is a significant difference to data collected by cycling apps like Strava (see below), whose users mainly pursue sporting goals.

The survey showed that the most important factors for most test persons in their choice of route were maximizing the use of cycle tracks and signposted routes, minimizing climbs and avoiding high traffic volumes and long waiting times. The findings gained from the GPS tracks confirmed this [ibid.]:

  • Only 19% of the trips were made on streets with high traffic volumes and without cyclist infrastructure. The majority of trips took place on streets with cyclist infrastructure or on side roads.
  • 27% longer trips were accepted in order to avoid one extra percent average incline.
  • To avoid non-signalized intersections, cyclists accepted a 13% longer trip.
  • 9% longer trips were accepted to avoid difficult left-turns.
  • Turning manoeuvres involve not only waiting times but also additional navigation. To avoid one additional turning manoeuvre, the cyclists were willing to take 1.5% longer trips.
  • Bridges had a great influence on the route choice, too. The test persons accepted on average 34% longer trips for bridges with separate cycle tracks and 20% longer trips for bridges with cycle lanes.

However, in general, cyclists, and commuters in particular, were very sensitive to circuitous routes. For trips up to 10 miles (16 kilometres), the median of the divergence between the route actually taken and the shortest route was only 400 meters; this corresponds to an additional expenditure of time of only 1.5 minutes [ibid.].


The Technical University of Graz and the University of Salzburg, together with the transport planning agency ZIS+P, the software manufacturer PTV and the app provider Bike Citizens, are currently developing, under a project named FamoS (bicycle traffic models as a planning instrument for reorganizing the street environment), a multimodal planning tool that will in particular better take account of the requirements of cyclists. The project will run until August 2018 and is funded by the Austrian Ministry for Transport, Innovation and Technology.


BikePrint, a Dutch planning software, was specifically developed for cycling planning. This programme accesses data collected during the Dutch Bicycle Count Week “Fiets Telweek” (see below) for its calculations.

The model makes it possible to calculate the impact of infrastructure projects on bicycle traffic, for example the travel-time reduction achieved by constructing a cycle superhighway. The programme can also simulate the impact of infrastructure measures on car traffic. In the Netherlands, cycling infrastructure measures are often considered to be the best measures for private motorized transport. If car drivers can be encouraged to switch to bicycles for short distances, this can also solve problems such as congestion and particulate matter.

Darstellung der Wartezeiten an Kreuzungen mit BikePrint
Darstellung der Wartezeiten an Kreuzungen mit BikePrint © Bussche 2016

The analysis of origin-destination pairs facilitates future network and infrastructure planning; accessibility maps show which sites can easily be reached by bicycle. These sites are perfect locations for facilities that attract large numbers of visitors, such as shopping centres or large cinema complexes. It is also possible to analyse travel-time losses at intersections. Comparing travel time and reference time identifies the origin-destination pairs on which cyclists experience delays.

BikePrint enables local authorities to check the usage of their main route network. Heat maps (see below) illustrate which routes are heavily used in practice. If these do not belong to the main route network, the local authority should try to improve the quality of the actual main route. Another option is of course to transform the heavily used alternative route into the main route, with all corresponding privileges such as priority in traffic light phasing or in winter maintenance.

A comparison of the actual number of cyclists with the fictitious number of those who would always take the shortest possible route can illustrate which routes are avoided by cyclists and which routes cause cyclists to even accept circuitous routes. This knowledge can help to make and justify investments in routes that are currently unimportant for bicycle traffic. However, if a relevant infrastructure is established, these routes could become important connections [Bussche 2016].

App crowdsourcing

Online route planners and app providers such as Naviki, Bike Citizens, Strava or Komoot collect movement patterns of their users. Naviki and Bike Citizens are primarily used for utility cycling, while Strava and Komoot tend to be aimed at a more sporting and leisure-oriented target group for training or outdoor activities. Apart from orientation and routing functions, tracking and statistics on the routes they travel are of particular interest for users. Another important aspect for cyclists is increasing their own “treasure of routes” (the amount of knowledge about routes they have travelled). The Bike Data Project website aims to collect data from users of the different apps worldwide and wants to make them available to planners. In addition, it offers its own app for tracking routes travelled. By the beginning of August 2017, more than 5 million kilometres had been tracked.

The data obtained make it possible to create so-called heat maps, which provide a fast and simple overview of the status quo of a city’s bicycle mobility. If the number of users is sufficient, heat maps can also be displayed for certain time intervals and can visualize the development over the course of the day in a time lapse.

Die BikeCitizens Heatmap von Berlin
Die BikeCitizens Heatmap von Berlin ©

INFORMATION BOX: heat maps in transport planning

A heat map is a diagram that visualizes data whose values are represented by colours. It serves to quickly and intuitively capture prominent values in a huge amount of data. [Oschabnig 2015]

Heat maps therefore make it possible to easily get an overview of the local traffic environment. Very busy routes are for example displayed by bold lines in dark red; rarely used routes are displayed by brighter thin lines. This enables transport planners to identify gaps in the cycle track network and to improve the cycling infrastructure of the local community and/or region [Reidl 2015]. But there are other factors such as stress and neuralgic cycling points that can be perfectly visualized by heat maps [Groß/Wilhelm 2014].

The data not only contain chosen routes but other important information for planners, for example average travel and dwell times, origin-destination matrices as well as deviations between the planned route and the route actually taken.

Data protection

The users are free to decide if they want to make their data available to the app providers in a pseudonymized form. To avoid the identification of individuals, Bike Citizens cuts off for example the first and last 100 m of every track. The company hands over the data it obtains only to local authorities and in an anonymized form; it does not sell data to the private sector.

Alongside use of the app, another motivation for the users to release their own data could be competition prizes, savings when buying certain products, the game effect or the idealistic contribution towards improving the cycling infrastructure.

App data for local authority planning

The major weakness and thus the main criticism as regards the use of app data is the question as to whether these data are representative. So far, most of the users have been technologically inclined young men, for portals like Koomot or Strava mostly focused on sporting activities. Another question is whether the apps are perhaps only used for unknown routes and not for daily, well-known routes.

An ongoing research project of the Federal Ministry of Transport and Digital Infrastructure, “Smartphone-generated behavioural data in cycling – usability assessment and development of evaluation guidelines for cycling planners”, conducted by the Technical University of Dresden, is examining this question. For this project, a random check of Strava’s GPS data was extrapolated to the actual level of traffic at the count sites of the city of Dresden. According to the research, a network-wide extrapolation with the help of count sites resulted in high-quality outcomes. Only the speeds of the Strava users were on average 5.5 km/h higher than the speeds actually measured at the count sites.

Interviews by the Technical University of Dresden showed that the main users are employed men aged 30 to 40 who use the app not only for exercise purposes but also for their everyday trips. The study concluded that app data can generally be used for cycling planning, but at least for Strava’s data the sporting background of the users is to be taken into account [Lißner et al. 2017].

In comparison to traditional data collection measures, the use of GPS-based app data has a very attractive price/quality ratio. Especially for local authorities that just started to promote cycling or that have already made some progress in promoting cycling, a good opportunity opens up to receive figures and data on local bicycle traffic and to implement effective cycling planning. This encourages the shift from supply-driven planning to demand-driven planning. The effects of cycling measures can easily be evaluated by comparing pre/post data [Francke et al. 2016]. The Technical University of Dresden is currently developing guidelines that provide local authorities and interested stakeholders with recommendations for the use of app data in practice.

The Fiets Telweek in the Netherlands

Since 2015, people in the Netherlands have been encouraged to participate in the annual Fiets Telweek (Bicycle Count Week). The participants download an app that tracks all routes cycled during the study period. The collected data give an overview of the population’s cycling behaviour and not only answer the question of the routes used by cyclists but also provide information on important points like the average speed of cyclists, the sites where there are often delays and the periods when most cyclists are on their bikes. These data give planners the opportunity to make bicycle traffic faster and to improve the network of cycle tracks. In 2015 and 2016, almost 50,000 persons took part in each Fiets Telweek. For the third run starting in September 2017, data will be collected throughout the whole year.

Die Heatmap der Fiets Telweek 2016
Die Heatmap der Fiets Telweek 2016 ©

Examples of further digital cycling planning tools

MVV cycling route planner

The analysis of traditional cycling route planners can provide interesting hints to planners, too. For an evaluation of the MVV cycling route planner, the most frequent origins and destinations and the resulting route calculations were examined. The results were used to present six potential cycle superhighways leading from Munich’s centre in all directions to the urban hinterland [Paul et al. 2016].


In 2013, the Technical University of Kaiserslautern analysed, within the context of a project named “EmoCycling”, situations of stress that cyclists face during their trips. The so-called “EmoMapping” then links emotions to geographic information. With the help of a smart bracelet, the research scientists captured situations of stress that the test persons had been facing and located them by GPS. In addition, the trips were recorded by a video camera. The results were then used to create heat maps that make dangerous spots in the city’s cycle network clearly visible. Further analyses made it possible to draw conclusions regarding the external factors that generate situations of stress at certain sites. Currently, the use of emotional city mapping for urban planning remains difficult as the original costs of procuring the equipment are very high and the evaluation is very time-consuming [Groß/Wilhelm 2014].

The Technical University of Kaiserslautern, the University of Heidelberg and the University of Salzburg are continuing to work on this subject in the follow-up project called “Urban Emotions”.


Within the framework of the “RadSpurenLeser” research project, which was funded by the Federal Ministry of Transport and Digital Infrastructure, the Innovation Centre for Mobility and Societal Change (InnoZ) used the “modalyzer” app to analyse the interface between bicycle and public transport. Between September 2015 and September 2016, the trips of 151 persons from Berlin and its surrounding area were tracked. The app was used as a digital trip diary and automatically mapped the travelled distances by the different means of transport. 12% of the bicycle trips travelled within the scope of the study were combined with public transport; in the course of the week, 78% of the participants had made multimodal transport operations. The collected data make it possible to draw further conclusions, for example on the catchment areas of train stations or on preferred transfer points. These findings can contribute to a more detailed planning of parking facilities or bike sharing schemes in the future [Howe et al. 2016].

PING if you care!

From June to November 2017, within the context of the “PING if you Care” pilot project, the city of Brussels, in cooperation with the Flemish association for sustainable mobility (Mobiel 21) and the app provider Bike Citizens, collected data on sites that cyclists consider unsafe. The participants, while on the move, were able to mark deficiencies such as pot-holes or poor visibility by a “PING button” that was connected to their smartphone via Bluetooth. They were subsequently able to categorize and comment on these points, including after they had got home. The project provided the city of Brussels with real-time information on the everyday situation of cyclists and enabled the users to actively participate in urban development. By the end of July 2017, more than 16,000 spots had already been identified.

Cycling reporting platform

The Cycling in Hesse reporting platform offers the possibility to report damage to cycling facilities and deficiencies compromising road safety. After choosing a local authority, a map section of the area concerned is displayed. The user can then mark a reported spot by directly entering an address, entering GPS coordinates or directly marking it on the map. For better orientation, the user can also switch to an aerial view. The next step involves demarcating the deficiency (e.g. surface, signposting, barriers, etc.) providing a more detailed description if needed. It is also possible to upload photos of the relevant location.

The participation of the cyclists ensures that existing deficiencies are detected in a fast and low-cost way. The competent entities in the local communities concerned mark the reported spots according to their degree of priority and enter the processing status in an internal area. The local authorities can also display all reports of their area and filter them by different criteria.

RADar! reporting platform

The Climate Alliance’s RADar! reporting platform offers participation opportunities for cyclists, too, and was developed within the context of the CITY CYCLING campaign. Users can mark a spot on a map and shortly describe the deficiency; the report is then sent to the local authority concerned. The local authority can mark the spot with different colours and thus display its processing status.

usschnitt aus dem RADar!-Stadtplan von Mainz, dargestellt werden Meldungen in Bearbeitung (gelb) und erledigte Meldungen (grün)
usschnitt aus dem RADar!-Stadtplan von Mainz, dargestellt werden Meldungen in Bearbeitung (gelb) und erledigte Meldungen (grün) ©

Deficiency Detective

The Saarland Tours App enables cyclists and hikers to report route damage or missing signs on the move via an integrated “deficiency detective”. The geographic data are transmitted to the Saarland tourist agency; a photo of the spot and a comment can be added, too.

Baden-Württemberg "Cycling to School" Travel Planer

Since the school year of 2016/2017, the Baden-Württemberg "Cycling to School" Travel Planer has enabled schoolchildren and parents to digitally record routes to school. Problematic spots can be marked and commented on. The findings inform the development of cycling to school travel plans and the local authority concerned can make danger spots more “forgiving”. This gives schoolchildren and their parents the opportunity to actively engage in the improvement of road safety.

Route to school check

With the help of the “route to school check” app, children and young people can actively participate in the creation of city maps for children and school travel plans. Alongside deficiencies, users can also mark interesting locations in local communities. Problem spots are checked and if necessary eliminated by local working groups from schools, the police and local authorities. All entries are displayed on the project’s website. Interested users can print out their own school or leisure map, tailored to their individual requirements [Leven 2014].

Conclusion and Looking Ahead

Digitalization is a great opportunity for cycling planning as well. Comprehensive data on cycling can be collected and evaluated with relatively simple means. But it is important to take into consideration that the collected data have to be seen in the context of the user group and should be converted to the total population if necessary. Simulating bicycle traffic with the help of traffic models will undoubtedly lead to more reliable results in the future, but programmes such as BikePrint or the recently launched Bike Citizens analytics tool already offer a good alternative today. In addition, cyclists can actively participate in cycling planning by providing their movement data in anonymized form or by using tools such as reporting platforms or cycling to school planners.

Checklist: use of digital tools for local authority planning

  • Traditional traffic models often ignore non-motorized traffic or do not sufficiently consider its requirements.
  • Permanent count sites can only selectively display a local community’s bicycle traffic; traffic counts are very costly and only display a short period.
  • App-generated data can be an opportunity to easily and cost-effectively gain area-wide data on local bicycle traffic.
  • Data generated by crowdsourcing should always be questioned with regard to their representativeness.
  • If a reporting platform is established, care should be taken to make available sufficient resources for processing the received reports. Non-processed reports will soon create frustration among users.


[Bussche 2016]
Bussche, Dirk (2016)
[Friderich et al. 2011]
in: Straßenverkehrstechnik 9/2011, S. 569-579
Friderich, Thomas, Joseph Broach, John Gliebe, Jennifer Dill (2011)
[Groß/Wilhelm 2014]
in: Planerin 4/2014, S. 63 f.
Groß, Dennis, Johann Wilhelm (2014)
[Howe et al. 2016]
in: Transforming Cities 4/2016, S. 68-71
Howe, Enrico, Robert Schönduwe, Andreas Graff, Lena Damrau, Joschka Kükenshöner (2016)
[Leven 2014]
in: mobilogisch 3/2014, S. 43 f.
Leven, Jens (2014)
[Lißner et al. 2017]
in: Internationales Verkehrswesen 1/2017, S. 48-52
Lißner, Sven, Angela Francke, Olena Chernyshova, Thilo Becker (2017)
[Oschabnig 2015]
in: mobilogisch 1/2015, S. 40
Oschabnig, Kerstin (2015)
[Paul et al. 2016]
in: Transportation Research Procedia 19/2016, S. 225-240 (Abruf am 09.08.2017)
Paul, Florian, Klaus Bogenberger, Bernhard Fink (2016)
[Reidl 2015]
in: Magazin 5/2015, S. 64-67
Reidl, Andrea (2015)
Meta Infos
SPT 09
Date (Text as of…)
28. August 2017
Handlungsfelder NRVP
Planning and developing a cycling strategy
Mobility behaviour