Changbai Wang

Faulting refers to the common and significant distress in Jointed Plain Concrete Pavement (JPCP), which has an adverse impact on the pavement roughness. Nevertheless, the existing fault prediction models continue to heavily rely on conventional linear regression techniques or basic machine learning approaches, which leaves room for improvement in training efficiency and interpretability. To enhance training efficiency and accuracy, this study developed five novel faulting prediction models. These models are based on five basic machine learning algorithms: Random Forest (RF), Additive Boosting (AdaBoost), Gradient Boosting Decision Tree (GBDT), Light Gradient Boosting Machine (LightGBM), and Categorical Boost (CatBoost), combined with the tree-structured Parzen estimator (TPE). The five models are TPE-RF, TPE-AdaBoost, TPE-GBDT, TPE-LightGBM, and TPE-CatBoost. In addition to selecting the best-performing model, this study incorporated the Shapley Additive Explanation (SHAP) technique and developed TPE-SHAP-CatBoost to improve the interpretability of the model’s predictions. The process involved extracting historical data on pavement performance, including 17 variables, from the Long-Term Pavement Performance (LTPP) database for 160 instances of observation. Firstly, the Boruta method was used to identify the final set of input variables. Secondly, the TPE technique, which is a Bayesian optimization method, was applied to automatically select the optimal hyperparameters for the base models. Finally, SHAP was used to provide both global and local explanations of the model’s outputs. The results indicate that the TPE-CatBoost model achieves the highest accuracy with an R² value of 0.906. Furthermore, the TPE-SHAP-CatBoost model identified the primary factors influencing faulting by incorporating SHAP and provided explanations of the model’s results at both the global and local levels. These research findings highlight the ability of the proposed model to accurately predict faulting, providing precise and interpretable guidance for pavement maintenance while reducing workload for pavement engineers in data collection and management.

Rutting is the main distress form of asphalt pavement, and its prediction accuracy is directly related to the reliability of the designed road. This research developed a neural network model to improve the prediction ability about the rutting of a pavement performance criterion and compared it with the multiple linear regression model and the existing neural network model. The neural network model is developed using the Keras module from the TensorFlow package in Python. Two reports generated by the National Cooperative Highway Research Program project 01-37A and the Long-Term Pavement Performance website records have been used as data sources for training the neural network model, which are reliable data preserved after years of monitoring. The input variables include the pavement thickness, service time, average annual daily traffic of trucks and the deformation of the asphalt concrete layer, granular base layer and subgrade layer. This experiment used 440 samples, of which 352 samples (80%) were used for model training and 88 samples (20%) for testing. The training results of the model reveal that the neural network model is significantly better than the multiple linear regression model, and the newly built neural network model performs better than another similar neural network in predictive performance. For the multiple linear regression model, the correlation coefficient R2 value between the measured and predicted in the testing set increased from 0.265 to 0.712. In contrast, it promotes from 0.867 to 0.902 for the neural network model.

The international roughness index (IRI) for roads is a crucial pavement design criterion in the Mechanistic-Empirical Pavement Design Guide (MEPDG). However, studies have shown that the IRI transfer function in the MEPDG is simply a linear combination of road parameters, so it cannot provide accurate predictions. To solve this issue, this research developed an AdaBoost regression (ABR) model to improve the prediction ability of IRI and compared it with the linear regression (LR) in MEPDG. The development of the ABR model is based on the Python programming language, using the 4265 records from the Long-Term Pavement Performance (LTPP) that include the pavement thickness, service age, average annual daily truck traffic (AADTT), gator cracks, etc., which are reliable data that are preserved after years of monitoring. The results reveal that the ABR model is significantly better than the LR because the correlation coefficient (R²) between the measured and predicted values in the testing set increased from 0.5118 to 0.9751, and the mean square error (MSE) decreased from 0.0245 to 0.0088. By analyzing the importance of variables, there are many additional crucial factors, such as raveling and bleeding, that affect IRI, which leads to the weak performance of the LR model.

To accelerate the resource utilization of coal gangue and meet the strategic requirements of carbon neutralization, alkali-activated, slag-cemented coal gangue is applied in the preparation of solid waste-based road stabilization materials. Here, the cementation characteristics and microstructure characteristics of alkali-activated, slag-cemented coal gangue road stabilization materials are studied using the alkali equivalent and coal gangue aggregate ratio as experimental variables. The results show that with the increase in alkali equivalent from 1% to 7%, the unconfined compressive strength of the alkali-activated coal gangue road stabilization material initially increases and then decreases, with 3% being the optimal group in terms of stabilization, the aggregate ratio of coal gangue increases from 70% to 85%, and the 7-day unconfined compressive strength of the stabilized material decreases approximately linearly from 8.16 to 1.68 MPa. At the same time, the porosity gradually increases but still meets the requirements of the specification. With the increase in hydration time, a large number of hydration products are formed in the alkali slag cementation system, and they are closely attached to the surface of and interweave with the coal gangue to fill the pores, resulting in the alkali slag slurry and coal gangue being brought closer together.

Under the modern and complicated urban environment, clarifying the interaction mechanism and mutual influence mechanism between the urban tunnel and the surrounding environment is of considerable significance to the construction and operation of the tunnel. To understand and grasp the excavation‐induced disturbance range and mechanism of adjacent pipeline jacking group on surrounding soils, this paper carries out relevant researches. Based on a case history of a three‐hole parallel adjacent jacking pipeline in shallow overburden, a field test and assessment are performed. By setting monitoring points around the monitoring section, the earth pressure and water pressure under different buried depths and different spacing are monitored. For further discussion, the general law of soil disturbance around single‐hole, double‐hole, and three‐hole jacking is investigated, the calculation method of lateral disturbance range and overlapping area range of pipe jacking group is deduced, and the degree of soil disturbance is evaluated as well. The results show that there will be different degrees of disturbance superposition in parallel multihole jacking due to the influence of the spacing, and the disturbance intensity is greatly affected by the parallel distance and buried depth.