MSc Thesis Presentation, SanthanaKrishanan Nayaranan


Review and modelling of shared autonomous vehicle services 28.06.2019 - 10.00

The actions of the autonomous vehicle manufacturers and related industrial partners as well as the interest from policy makers and researchers point towards the initial deployment of autonomous vehicles as a shared autonomous mobility service. Numerous research stud- ies are lately being published regarding Shared Autonomous Vehicles (SAVs) and hence it is necessary to have a comprehensive outlook to consolidate the existing knowledge base. Hence, one of the twofold objectives of this thesis is to present a comprehensive consolida- tion of studies in the field of SAV services. The primary focus is the critical evaluation of the impacts which are categorised into seven groups namely Travel behaviour, Traffic, Transport Supply, Land-use, Economy, Environment and Governance. Pertinent to the evaluation of the impacts, an SAV typology is presented and the modelling approaches, expected demand and policy framework required are reviewed. The second research objective of this thesis is to formulate and solve combined Dynamic User Equilibrium and SAV Chain Formation (DUESCF) problem as a bilevel model based on Nash-Cournot game involving road users and SAV service operator, with the assumption that the road network in future is going to be filled with conventional private vehicles and reservation based shared autonomous vehicles. In such a scenario, road users select paths and departure times to minimize their disutility forming a DUE (DVI – fixed point problem) and SAV service operator try to maximize their performance forming appropriate SAV chains (combinatorial problem). The final objective of this formulation is a traffic assignment and SAV chain formation, such that both road users and SAV service operator obtain their optimal solutions simultaneously, forming a Nash-Cournot equilibrium where no player is better off by unilaterally changing their deci- sions. A solution approach based on Iterative Optimization and Assignment (IOA) method is proposed with path flow and SAV performance changes as convergence criteria. Further, the solution approach is tested for its robustness using an existing DUE and SAV chain formation model from the literature and the tests are done on three different networks with varying level of complexity. This is the first time such a model is being formulated and solved in literature.