Search

Due to the irregular features of the airport pavement cracks with variable shapes，narrow widths， different lengths，and free⁃form spatial trends，the existing algorithms perform not well. To solve this problem，this paper constructs a neural network model based on deformable convolution and feature fusion （DFNet）. Firstly， the deformable convolution module is used to enhance the learning of fracture morphological characteristics by feature extraction network. Secondly，the multi⁃scale convolution module captures the global information of the fracture under different receptive fields. Finally，the feature fusion module is used to extract the characteristics of different levels of the fracture，through combining low⁃level and high⁃level features of cracks to achieve accurate segmentation of cracks on airport pavements. On the collected actual airport pavement crack data set，comparative experiments are carried out with other six existing algorithms. The proposed algorithm reaches 90.95% on the F1⁃Score of pixel⁃level segmentation， which is better than those of all other compared algorithms. The DFNet algorithm improves the ability of detecting cracks in airport pavements. Experimental results show that the proposed algorithm better meets the actual engineering requirements. [ABSTRACT FROM AUTHOR]

Dispatching rule is an effective method for solving dynamic Job-Shop scheduling problem in practical production. However, its biggest problem lies in that usually it only has good performance in the specific scheduling environment, so real-time selection and evaluation are needed. To study the real-time scheduling problem with dynamic shop in practical production, this paper surveyed the methods for the selection and evaluation of dispatching rules. It reviewed the development, classification and characteristics of dispatching rules, and summarized the research hotspots of dispatching rules including the selection and evaluation methods. It introduced the selection methods of dispatching rules in detail which included the popular steady state simulation method and the effective artificial intelligence method. In addition, it presented the research results and conclusions of simulation methods, expert system, machine learning methods and artificial neural network methods that were applied to the selection of dispatching rules. Besides, it introduced the measures and methods for the evaluation of dispatching rules. Finally this paper pointed out the direction of future research aiming at the shortcomings of the existing dispatching rules. [ABSTRACT FROM AUTHOR]