About the project
The aim of this project is to leverage the potential of big data and AI to obtain a clearer insight into effects such as vehicle technical characteristics and driver/operator moods, preferences, and behaviours, and use these insights to improve existing operational and strategic policies. Relevant data from existing large vehicle fleets will be utilised to develop models that will be integrated into a prototype training platform to be used across different fleet operators in the UK and internationally.
Road-based vehicle fleets are the cornerstone of modern-day transport and logistics systems, supporting a wide range of passenger and freight travel needs. From public buses and Heavy Goods Vehicles (HGVs) operating in interurban environments to taxis and cargo cycles serving dense city cores, fleets represent a sizeable proportion of traffic on the roads, and can therefore be attributed a considerable share of the resulting adverse impacts, such as congestion, accidents, energy consumption, pollution, and noise.
Addressing these impacts at the source, i.e., at the individual vehicle and driver/operator level, can, therefore, deliver substantial benefits for the whole of the transport system. Such an endeavour, however, has not been fruitful to date due to a prevailing lack of methods and tools aimed at understanding the effects of different vehicle- and driver-related parameters on the efficiency, safety, and sustainability of fleet operations.
The models will integrate optimisation under uncertainty ( e.g., Markov decision processes), preference/choice modelling analytics and operational research to embed meaningful decision support into the training platform and derive useful insights from the dataset.