Traffic signal control and route choice, capacity maximization and stability
Event details
| Date | 06.03.2015 |
| Hour | 12:15 › 13:15 |
| Speaker | Prof. Dr Mike Smith, University of York, England |
| Location | |
| Category | Conferences - Seminars |
The talk will have two parts: A, B.
PART A: [Routeing + control] dynamics with just flows and green-times.
This talk outlines the P0 control policy in a model without queues. It is shown that under certain conditions this policy maximises network capacity.
Natural dynamical flow-only models of day-to-day re-routeing (drivers swap to quicker routes) is combined with day to day green-time response based on the responsive control policy P0 (see Smith (1979, 1980, 1987). It is shown that, for any steady feasible demand within this dynamical flow + control model, any (flow, green-time) solution trajectory cannot leave the region of supply-feasible (flow, green-time) pairs, costs are bounded, and throughput is maximised.
PART B: [Routeing + control] dynamics with flows and green-times and queues.
It will then be shown that, within a model with explicit queues, modified Varaiya (2013) and Le at al (2013) control policies do not have the above properties. We give a network and a steady demand within the capacity of the network for which there is no equilibrium consistent with either policy; in each case a dynamical [routeing + control] model has unbounded queues.
The paper shows that, in contrast, responsive P0 on this example network does give rise to a consistent equilibrium and a dynamical [routeing + control] model has bounded queues. (One possible extension of P0 ideas to include Macroscopic Fundamental Diagrams may be described briefly).
Bio :
Mike is Emeritus Professor of Mathematics at the University of York; he was awarded the Robert Herman Lifetime Achievement Award in 2007 for his work on the Mathematics of Transport.
Mike entered the University of York as a lecturer in Mathematics in 1964. Soon after, the construction of a wide ring road close to York City Centre (and the demolition of 450 houses) was proposed to reduce congestion in York.
Mike’s first paper (“Traffic control in a town with two ring roads”) was published in 1974. This showed how traffic control could be used to reduce congestion in York without constructing a wide inner ring and without demolishing houses.
In 1979 / 1980 Mike published a paper called “A Local Traffic Control Policy which Automatically Maximises the Overall Travel Capacity of a Road Network”. Since then he has spent many years extending this “magic” policy (called P0); showing how traffic control may be used to reduce congestion within an existing road network; while taking account of route choices.
PART A: [Routeing + control] dynamics with just flows and green-times.
This talk outlines the P0 control policy in a model without queues. It is shown that under certain conditions this policy maximises network capacity.
Natural dynamical flow-only models of day-to-day re-routeing (drivers swap to quicker routes) is combined with day to day green-time response based on the responsive control policy P0 (see Smith (1979, 1980, 1987). It is shown that, for any steady feasible demand within this dynamical flow + control model, any (flow, green-time) solution trajectory cannot leave the region of supply-feasible (flow, green-time) pairs, costs are bounded, and throughput is maximised.
PART B: [Routeing + control] dynamics with flows and green-times and queues.
It will then be shown that, within a model with explicit queues, modified Varaiya (2013) and Le at al (2013) control policies do not have the above properties. We give a network and a steady demand within the capacity of the network for which there is no equilibrium consistent with either policy; in each case a dynamical [routeing + control] model has unbounded queues.
The paper shows that, in contrast, responsive P0 on this example network does give rise to a consistent equilibrium and a dynamical [routeing + control] model has bounded queues. (One possible extension of P0 ideas to include Macroscopic Fundamental Diagrams may be described briefly).
Bio :
Mike is Emeritus Professor of Mathematics at the University of York; he was awarded the Robert Herman Lifetime Achievement Award in 2007 for his work on the Mathematics of Transport.
Mike entered the University of York as a lecturer in Mathematics in 1964. Soon after, the construction of a wide ring road close to York City Centre (and the demolition of 450 houses) was proposed to reduce congestion in York.
Mike’s first paper (“Traffic control in a town with two ring roads”) was published in 1974. This showed how traffic control could be used to reduce congestion in York without constructing a wide inner ring and without demolishing houses.
In 1979 / 1980 Mike published a paper called “A Local Traffic Control Policy which Automatically Maximises the Overall Travel Capacity of a Road Network”. Since then he has spent many years extending this “magic” policy (called P0); showing how traffic control may be used to reduce congestion within an existing road network; while taking account of route choices.
Practical information
- General public
- Free
Organizer
- Prof. Dr Nikolas Geroliminis & Katrin Beyer
Contact
- Prof. Dr Nikolas Geroliminis