TAPAS

on YouTube ↗34m

" summary

An open source agent based demand model generates address level daily activity chains and feeds them into SUMO for microscopic routing and traffic assignment, using a synthesizer to create households and vehicles and extra commercial traffic. A Berlin study runs about 3.5 million agents with billions of potential routes collapsed by task grouping to avoid 4.5 trillion routings. Iterative feedback produces stable travel times with equilibrium after three iterations and median deviation near zero. Example uses include public transport pricing, demand responsive transport and city scale air quality measures.

AI-written summary · about

" chapters (16)

1. 0:00Intro
2. 1:08Outline
3. 1:48Background: Travel demand modelling
4. 3:34TAPAS-SUMO: Principle of operation
5. 5:52Objectives and und applications of the region model TAPAS (Travel-activity pattern simulation)
6. 7:50TAPAS: Simulation, Input, Output
7. 9:29Data preparation
8. 12:12TAPAS: Generation
9. 12:36TAPAS: Empiric activity plans
10. 15:02TAPAS: Destination choice
11. 19:14TAPAS: Mode choice via multinomial logit
12. 23:10TAPAS: Why traffic assignment zones?
13. 25:06TAPAS triptransfer
14. 27:04SUMO workflow
15. 29:23Results: Time deviation of different iterations
16. 31:33Results: Traffic flow on certain streets