As bicycling becomes an integral part of sustainable mobility, reliable planning tools are essential to ensure bicycling as an efficient mode of transport. The growing bicycle demand requires not only expanding infrastructure, but also ensuring that such infrastructure supports well-functioning traffic under high demands. Given the high heterogeneity in bicyclist characteristics, the use of microscopic traffic simulation, which explicitly considers individual properties and preferences, becomes particularly useful for evaluating bicycle traffic performance. While traffic simulation has been extensively utilized for traffic planning of various modes of transport, this type of modeling support is largely lacking in the planning of bicycle traffic. Although most commercial simulators allow multi-modal traffic analysis, bicycle traffic is often modeled by adjusting parameters in models originally designed for other modes, even though bicyclists may exhibit distinct characteristics and behaviors. Consequently, the proper inclusion of bicyclists into various traffic simulation analyses is difficult, and often inaccurate. The objective of this thesis is to develop and evaluate mathematical models for accurate microscopic simulation of bicycle traffic, with a focus on developing empirically well-founded models that capture the heterogeneity in bicyclists’ characteristics and preferences, as well as their interactions with the built environment and with each other. The thesis delivers an empirical characterization of bicycle traffic in diverse contexts, describing the heterogeneity in characteristics and preferences of bicyclists—including disaggregated analyses by bicycle type— that potentially influence traffic performance. Methods for processing and validating bicycling data are developed to support this characterization. Furthermore, the thesis demonstrates that bicyclist speeds are highly context-dependent and proposes simulation models for context-related features of bicycling trips, such as topography, curvature, and wind, that integrate heterogeneous and adaptive free riding behavior to improve the accuracy of simulated speeds and the reliability of bicycle traffic simulations. This thesis advances the accuracy and applicability of microscopic simulation of bicycle traffic for its use in the planning of well-functioning bicycle traffic.

Cykling är en viktig del av hållbar mobilitet, men för att säkerställa cykling som ett effektivt transportslag krävs pålitliga verktyg för planering och beslut. Den växande efterfrågan på cykling kräver inte bara utbyggd infrastruktur, utan också att infrastrukturen möjliggör välfungerande trafik vid höga belastningar. Givet den stora variationen i cyklisters egenskaper blir mikroskopisk trafiksimulering, som tar hänsyn till individuella egenskaper och preferenser, särskilt användbar för att utvärdera cykeltrafikens framkomlighet. Samtidigt som trafiksimulering länge har använts inom planering för olika transportslag saknas denna typ av verktyg i stor utsträckning för cykeltrafik. Även om de flesta kommersiella simulatorer kan analysera flera trafikslag modelleras cyklister ofta genom att justera parametrar i modeller som ursprungligen designades för andra transportslag, trots att cyklister ofta upp-visar unika egenskaper och beteenden. Det gör det svårt att få korrekta resultat när man simulerar cykeltrafik. Syftet med denna avhandling är att utveckla och utvärdera matematiska modeller för noggrann mikroskopisk simulering av cykeltrafik, med fokus på modeller som fångar egenskaper och preferenser hos cyklister, liksom deras interaktioner med infrastrukturen och med varandra. Avhandlingen beskriver cykeltrafik i olika sammanhang, inklusive hur egenskaper och preferenser varierar mellan olika cykeltyper, vilket kan på-verka trafikens framkomlighet. Detta baseras på metoder för bearbetning och validering av cykeldata som utvecklades i avhandlingen. Vidare föreslår avhandlingen simuleringsmodeller för hur olika förhållanden på en cykeltur påverkar cyklister, såsom lutningar, kurvor, och vind. Modellerna tar hänsyn till att cyklister är olika och anpassar sig till förhållandena, vilket förbättrar noggrannheten och tillförlitligheten i simuleringarna. Denna avhandling bidrar till ökad användbarhet hos mikroskopisk simulering av cykeltrafik för att planera välfungerande cykeltrafik.

To simulate bicycle traffic accurately, it is essential to capture how bicyclists react to features of the infrastructure such as the longitudinal gradient of a bicycle path. Bicycling requires human-powered motion, and the power output provided by bicyclists differ significantly among bicyclists due to physical capabilities and preferences. Therefore, the objective of this paper is to investigate the connection between power output and gradient in bicycle traffic, with the purpose of developing a power-based model that predicts accurately the speed of bicyclists. Based on trajectory data of free-riding bicyclists travelling on a non-flat bicycle path segment, we estimate changes in power output as a function of gradient considering the physical forces acting on a bicycle. The results suggest a linear correlation between power output and gradient; while bicyclists increase their power output on the uphill as gradient increases, they decrease their power output on the downhill as gradient increases. By implementing this correlation into a traffic simulation algorithm, we show that the simulation captures well the impact of gradients in a population of bicyclists as it reproduces similar speed profiles. We conclude that bicyclists adapt their power output to compensate for the gradient and its associated change in speed, and that the impact of gradient varies greatly among bicyclists. Furthermore, we conclude that power-based modelling of free-riding bicyclists is an attractive alternative to investigate further.

As bicycling becomes an integral part of sustainable mobility, it becomes essential to enhance planning strategies that ensure bicycling as an efficient mode of transport. While traffic simulation has been extensively utilized for traffic planning of various modes of transport, this type of modeling support is largely lacking in the planning of bicycle traffic.

Given the high heterogeneity in the characteristics of bicyclists, the use of microscopic traffic simulation, which incorporates the explicit inclusion of individual properties and preferences, becomes particularly useful for evaluating bicycle traffic performance.

By examining real-world traffic, the objective of this thesis is to investigate essential requirements for microscopic modeling and simulation of bicycle traffic on off-street bicycle path segments, and to further develop and evaluate modeling approaches suitable for bicycle traffic. Understanding the fundamentals of how bicyclists interact with the infrastructure and other bicyclists is a necessary step towards accurate simulation of bicycle traffic.

In this thesis, research gaps related to the evaluation of bicycle traffic performance and simulation are identified, and methods to validate bicycling data are proposed to determine its quality and suitability for traffic analysis. Furthermore, two distinct modeling approaches are investigated to simulate the impact of gradients in bicycle traffic. The first involves calibrating a car-based model using a widely-used microscopic traffic simulation software, and the second implements a power-based model rooted in the physical forces acting on a bicycle. Lastly, characteristics of bicycle traffic that are relevant for simulating bidirectional traffic are identified and described.

The work in this thesis offers a starting point towards enhanced microscopic bicycle traffic simulation that effectively assist the planning of efficient bicycle traffic.