A Comparative Study of AI-Based International Roughness Index (IRI) Prediction Models for Jointed Plain Concrete Pavement (JPCP)

Wang, Qiang; Zhou, Mengmeng; Sabri, Mohanad Muayad Sabri; Huang, Jiandong

doi:10.3390/ma15165605

Cited by 4 publications

(1 citation statement)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reduce the risk of trapping in local optima, the step size of the beetle can be adjusted by Levy flight and self-inertia weight. The improved BAS has been used in the prediction of concrete properties and the design of concrete mix in past studies [ 105 , 106 , 107 , 108 , 109 , 110 ]. It should be noted that for traditional BAS used in concrete design, the beetle step size is constant or decreases in each iteration.…”

Section: Future Prospectsmentioning

confidence: 99%

Mixture Optimization of Cementitious Materials Using Machine Learning and Metaheuristic Algorithms: State of the Art and Future Prospects

Song¹,

Wang

Li³

et al. 2022

Materials

Self Cite

View full text Add to dashboard Cite

The hybrid optimization of modern cementitious materials requires concrete to meet many competing objectives (e.g., mechanical properties, cost, workability, environmental requirements, and durability). This paper reviews the current literature on optimizing mixing ratios using machine learning and metaheuristic optimization algorithms based on past studies on varying methods. In this review, we first discuss the conventional methods for mixing optimization of cementitious materials. Then, the problem expression of hybrid optimization is discussed, including decision variables, constraints, machine learning algorithms for modeling objectives, and metaheuristic optimization algorithms for searching the best mixture ratio. Finally, we explore the development prospects of this field, including, expanding the database by combining field data, considering more influencing variables, and considering more competitive targets in the production of functional cemented materials. In addition, to overcome the limitation of the swarm intelligence-based multi-objective optimization (MOO) algorithm in hybrid optimization, this paper proposes a new MOO algorithm based on individual intelligence (multi-objective beetle antenna search algorithm). The development of computationally efficient robust MOO models will continue to make progress in the field of hybrid optimization. This review is adapted for engineers and researchers who want to optimize the mixture proportions of cementitious materials using machine learning and metaheuristic algorithms.

show abstract

Section: Future Prospectsmentioning

confidence: 99%

Mixture Optimization of Cementitious Materials Using Machine Learning and Metaheuristic Algorithms: State of the Art and Future Prospects

Song¹,

Wang

Li³

et al. 2022

Materials

Self Cite

View full text Add to dashboard Cite

show abstract

Developing an improved extreme gradient boosting model for predicting the international roughness index of rigid pavement

Wang,

Xiao,

Liu

2023

Construction and Building Materials

View full text Add to dashboard Cite

Roughness Prediction of Jointed Plain Concrete Pavement Using Physics Informed Neural Networks

Pasupunuri,

Thom,

2024

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

The International Roughness Index is used to measure the road roughness in pavements, as pavement roughness deteriorates over time. Despite many attempts by researchers to predict roughness in concrete pavements, there are limitations, such as small sample size, modeling approach, or lack of robustness in the model. This study presents a novel machine-learning approach incorporating domain knowledge to predict roughness, using a dataset obtained from the Long-Term Pavement Performance database. Physics informed neural networks (PINNs) are popular physics-driven machine-learning approaches that have been receiving widespread attention in the field of civil engineering. PINNs work similarly to neural networks but are augmented with the incorporation of physics-based constraints and governing equations, enabling them to assimilate domain knowledge and leverage physical principles while making predictions or solving problems. In this study, the popular Mechanistic-Empirical Pavement Design Guide roughness prediction model is used along with the optimized neural networks to calculate the physics-based loss function. The Optuna framework is used to tune the hyperparameters within the neural network architecture. The final configuration, optimized and trained in the model, has three hidden layers with, respectively, 27, 67, and 80 neurons. The tuned model has performed well for the testing dataset, with a mean absolute error of 0.134 and a coefficient of determination of 0.90. A sensitivity analysis was also conducted and is presented to understand the influence of the variation of each variable.

show abstract

A Comparative Study of AI-Based International Roughness Index (IRI) Prediction Models for Jointed Plain Concrete Pavement (JPCP)

Cited by 4 publications

References 57 publications

Mixture Optimization of Cementitious Materials Using Machine Learning and Metaheuristic Algorithms: State of the Art and Future Prospects

Mixture Optimization of Cementitious Materials Using Machine Learning and Metaheuristic Algorithms: State of the Art and Future Prospects

Developing an improved extreme gradient boosting model for predicting the international roughness index of rigid pavement

Roughness Prediction of Jointed Plain Concrete Pavement Using Physics Informed Neural Networks

Contact Info

Product

Resources

About