A multi-task deep learning approach for lane-level pavement performance prediction with segment-level data

TL;DR

This paper introduces a multi-task deep learning model using LSTM layers to predict lane-level pavement performance from segment-level data, improving accuracy and capturing lane-specific differences.

Contribution

It presents a novel multi-task LSTM-based framework that effectively models lane-level pavement performance using segment-level data, addressing data collection challenges.

Findings

Achieved less than 10% mean absolute percentage error in predictions.

Outperformed ensemble and shallow machine learning methods.

Validated with real case data in China.

Abstract

The elaborate pavement performance prediction is an important premise of implementing preventive maintenance. Our survey reveals that in practice, the pavement performance is usually measured at segment-level, where an unique performance value is obtained for all lanes within one segment of 1km length. It still lacks more elaborate performance analysis at lane-level due to costly data collection and difficulty in prediction modeling. Therefore, this study developed a multi-task deep learning approach to predict the lane-level pavement performance with a large amount of historical segment-level performance measurement data. The unified prediction framework can effectively address inherent correlation and differences across lanes. In specific, the prediction framework firstly employed an Long Short-Term Memory (LSTM) layer to capture the segment-level pavement deterioration pattern. Then multiple task-specific LSTM layers were designed based on number of lanes to capture lane-level differences in pavement performance. Finally, we concatenated multiple task-specific LSTM outputs with auxiliary features for further training and obtained the lane-level predictions after fully connected layer. The aforementioned prediction framework was validated with a real case in China. It revealed a better model performance regardless of one-way 2-lane, 3-lane, and 4-lane scenarios, all lower than 10% in terms of mean absolute percentage error. The proposed prediction framework also outperforms other ensemble learning and shallow machine learning methods in almost every lane.