Välkommen till Transportbibliotekets katalog

Normalvy MARC-vy

Application of stochastic learning automata for modeling departure time and route choice behavior Ozbay, Kaan ; Datta, Aleek ; Kachroo, Pushkin

Av: Medverkande: Utgivningsinformation: Transportation Research Record, 2002Beskrivning: nr 1807, s. 154-62Ämnen: Bibl.nr: VTI P8167:1807Location: Abstrakt: Stochastic learning automata (SLA) theory is used to model the learning behavior of commuters within the context of the combined departure time route choice (CDTRC) problem. The SLA model uses a reinforcement scheme to model the learning behavior of drivers. A multiaction linear reward-e-penalty reinforcement scheme was introduced to model the learning behavior of travelers based on past departure time choice and route choice. A traffic simulation was developed to test the model. The results of the simulation are intended to show that drivers learn the best CDTRC option, and the network achieves user equilibrium in the long run. Results indicate that the developed SLA model accurately portrays the learning behavior of drivers, while the network satisfies user equilibrium conditions.
Exemplartyp: Rapport, konferenser, monografier
Bestånd
Aktuellt bibliotek Status
Statens väg- och transportforskningsinstitut Tillgänglig

Stochastic learning automata (SLA) theory is used to model the learning behavior of commuters within the context of the combined departure time route choice (CDTRC) problem. The SLA model uses a reinforcement scheme to model the learning behavior of drivers. A multiaction linear reward-e-penalty reinforcement scheme was introduced to model the learning behavior of travelers based on past departure time choice and route choice. A traffic simulation was developed to test the model. The results of the simulation are intended to show that drivers learn the best CDTRC option, and the network achieves user equilibrium in the long run. Results indicate that the developed SLA model accurately portrays the learning behavior of drivers, while the network satisfies user equilibrium conditions.