There has been a recent trend of cooperation between local traffic authorities and in-vehicle
route guidance service providers that involves free exchange of traffic data. It is important for
both sides to understand the outcomes of the cooperation and how to improve their
individual outcomes. Despite the relevance of this topic, this area has not been explored in
detail by existing studies. Hence, this thesis focuses on investigating cooperation strategies
for both local authorities and service providers.
This research first develops a cooperation strategy evaluation framework for free data
exchange schemes. The framework consists of two model components. The first component
is a mixed Path Size Logit (PSL) based market equilibrium model that estimates the market
response to a given cooperation strategy. The second component comprises functions to
evaluate the benefit for local traffic authorities and service providers. In particular, the benefit
functions for free service providers and paid service providers are developed based on
assumptions about their differences in revenue sources.
Next, a framework for optimising cooperation strategies for the local authority and service
providers is developed based on a non-cooperative Nash game. The framework is
formulated as an Equilibrium Problem with Equilibrium Constraints (EPEC), which is solved
by a Particle Swarm Optimiser (PSO) based Diagonalization algorithm that is developed as a
part of this research. This optimisation framework is subsequently extended to analyse
cooperation scenarios involving paid data sharing schemes and a mix of free and paid
schemes.
A number of numerical studies are conducted using the developed frameworks. One
numerical study investigates the impact of market characteristics on the market share of a
service provider with relatively good data quality, and finds that the route choice set
composition, drivers’ value of time (VOT) and their perception accuracy with respect to
service providers’ utility have a significant impact on the market share. The other numerical
studies show three general implications for the cooperation. Firstly, when a local traffic
authority’s objective is to maximise the network performance, it should cooperate with all
service providers via free data exchange schemes and make sure that cooperative service
providers use the data to enhance their data quality and not just reduce their own data
acquisition costs. Secondly, when a local traffic authority considers generating a revenue by
selling its data to cooperative service providers (via paid schemes), the authority should be
aware that its revenue does not always increase when the data is sold to more service
providers; higher revenues can sometimes be obtained from service providers who get a
data quality advantage over other providers from local authority data. Finally, there may be a
public concern about whether the local traffic authority’s data is appropriately utilised for
improving the public benefit when the local traffic authority cooperates with service providers
via paid schemes, and such concerns may prevent the local traffic authority from engaging in
commercial transactions with service providers through paid data exchange schemes.