Deep learning-based optimal selection of construction and demolition waste crushing equipment

The crushing equipment plays a crucial role in the recycling process of construction and demolition waste (CDW). Not only does it have a significant impact on the performance of recycled aggregates, but also influences the costs and environmental emissions associated with CDW recycling. To address the limitations of the traditional manual selection method, this study proposed an optimal selection method for CDW crushing equipment based on deep learning. Firstly, a rapid assessment method for the size and volume of CDW based on deep learning was proposed. Specifically, the Mask R-CNN model was employed to identify and segment the CDW at the demolition site. Thereafter, the size distribution and mass distribution of the CDW were calculated based on the segmentation results, using the Brute Force algorithm and the 3D volume reconstruction method, respectively. Secondly, the size distribution of the CDW was combined with the discharge size requirement for recycled aggregates to determine the optional crushing equipment that can meet the CDW production requirements. Finally, the environmental emissions of optional crushing equipment were calculated based on life cycle assessment (LCA) and converted to environmental costs based on the social willingness to pay (WTP). Subsequently the optimal crushing equipment for CDW was determined by combining running costs and environmental costs. The results demonstrated that the Mask R-CNN model employed in this study exhibited superior accuracy in comparison to other segmentation models. The overall error in the size distribution and mass distribution was maintained within 5 %. The method of this study can determine the optimal crushing equipment for CDW and provide auxiliary decision support for actual recycling projects. Furthermore, the crushing equipment selected by this research method can effectively control the environmental emissions during the CDW crushing stage, thereby facilitating the low-carbon recycling of CDW, which is conducive to the sustainable development of the construction industry.