Anita Graser, Maximilian Leodolter and Hannes Koller
Towards Better Urban Travel Time Estimates Using Street Network Centrality
http://eurocarto.org/wp-content/uploads/2015/10/2-13.pdf
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Better Urban Travel Time Estimates Using Street Network Centrality #Eurocarto15
1. Towards Better Urban Travel Time
Estimates Using Street Network Centrality
Anita Graser, Maximilian Leodolter & Hannes Koller
2. Problem Statement
We want to:
1. Predict travel time / speed on a certain road at a certain time
2. Where we don’t have measurements
3. Using only street network data – NO travel demand data
2
3. Travel Time Prediction
Usual approaches:
Sensor measurements
Traffic model
Fallback: Speed limit ×correction factor
3
4. 4
Our Basic Prediction Model
MAPE
Explanatory
Variables:
- Time of day
- Road class
- Speed limit
5. Centrality
1. Closeness: “Central” “Peripheral”
How far is it to all other network nodes?
2. Betweenness: “Important” “Unimportant”
How often is this location on the shortest paths between network nodes?
5
Die genauen grenzen für die definition von central, important,…:
[0,0.333] rural bzw. Not important
(0.333,0.667] avg. rural bzw. Avg. important
(0.667,1] central bzw. important
Mape for: Cuf == 2, cc == 1, (base 0.236), (global 0.216), (lokal 0.203)
Improvement base -> local ca. 14%
Improvement base -> global ca. 8.5%
Improvement global -> local ca. 6%