From: ASENN: attention-based selective embedding neural networks for road distress prediction
Input variables | Value | |
---|---|---|
Pavement data | Road type | Arterial |
Maintenance type | Partial | |
Age of the road section | 1 | |
Year of data collection | 2015 | |
Environmental data | Temperature | 26 |
Humidity | 49 | |
Atmospheric pressure | 1019 | |
Traffic data | Traffic count (heavy vehicles) | 4,203,429 |
Traffic count (light vehicles) | 487,421 | |
Direction of traffic | Forward | |
Road distress parameter | Cracking at t−1 | 0 |
Deflection at t−1 | 49 | |
IRI at t−1 | 1 | |
Rutting at t−1 | 2.674 | |
Label (Output) | Value | |
Road distress parameter | Cracking at t | 0 |
Deflection at t | 58.8 | |
IRI at t | 1.2 | |
Rutting at t | 3.07625 |