CODD: A benchmark dataset for the automated sorting of construction and demolition waste.

This study presents the Construction and Demolition Waste Object Detection Dataset (CODD), a benchmark dataset specifically curated for the training of object detection models and the full-scale implementation of automated sorting of Construction and Demolition Waste (CDW). The CODD encompasses a comprehensive range of CDW scenarios, capturing a diverse array of debris and waste materials frequently encountered in real-world construction and demolition sites. A noteworthy feature of the presented study is the ongoing collaborative nature of the dataset, which invites contributions from the scientific community, ensuring its perpetual improvement and adaptability to emerging research and practical requirements. Building upon the benchmark dataset, an advanced object detection model based on the latest bounding box and instance segmentation YOLOV8 architecture is developed to establish a baseline performance for future comparisons. The CODD benchmark dataset, along with the baseline model, provides a reliable reference for comprehensive comparisons and objective assessments of future models, contributing to progressive advancements and collaborative research in the field.

Real-time construction demolition waste detection using state-of-the-art deep learning methods; single-stage vs two-stage detectors.

Central to the development of a successful waste sorting robot lies an accurate and fast object detection system. This study assesses the performance of the most representative deep-learning models for the real-time localisation and classification of Construction and Demolition Waste (CDW). For the investigation, both single-stage (SSD, YOLO) and two-stage (Faster-RCNN) detector architectures coupled with various backbone feature extractors (ResNet, MobileNetV2, efficientDet) were considered. A total of 18 models of variable depth were trained and tested on the first openly accessible CDW dataset developed by the authors of this study. This dataset consists of images of 6600 samples of CDW belonging to three object categories: brick, concrete, and tile. For an in-depth examination of the performance of the developed models under working conditions, two testing datasets containing normally and heavily stacked and adhered samples of CDW were developed. A comprehensive comparison between the different models yields that the latest version of the YOLO series (YoloV7) attains the best accuracy (mAP50:95 ≈ 70%) at the highest inference speed (<30 ms), while also exhibiting enough precision to deal with severely stacked and adhered samples of CDW. Additionally, it was observed that despite the rising popularity of single-stage detectors, apart from YoloV7, Faster-RCNN models remain the most robust in terms of exhibiting the least mAP fluctuations over the testing datasets considered.

Development of a High Strength Geopolymer Incorporating Quarry Waste Diabase Mud (DM) and Ground Granulated Blast-Furnace Slag (GGBS).

This study presents the development and experimental assessment of novel, high strength, cementless binders that incorporate alkali-activated local waste. A silica-rich diabase mud (DM), currently considered as waste, was previously investigated for geopolymerization, signifying that the DM lacked the necessary reactivity to provide a stable geopolymer binder alone. Moreover, even after incorporation of small amounts of cement and metakaolin, the DM mixtures still did not yield adequate mechanical properties. In this study, the local DM was instead combined with another industrial byproduct known as Ground Granulated Blast-furnace Slag (GGBS) in varying mixtures. The mixture design trials enabled the development of three high strength cementless geopolymer mixtures with 28-day compressive strengths ranging between 60 and 100 MPa, comparable to conventional concrete compressive strengths. The results indicate that the innovative geopolymer material is very promising for the manufacturing of pavement tiles and other precast construction products. Most importantly, this study presents the first successful development of a construction material of adequate compressive strength that can absorb large quantities of the abundant quarry waste, following a course of 10 years of unsuccessful attempts to valorize the local DM. Although difficulties were encountered due to a high reactivity rate, especially for the mix that included the highest GGBS content, prototype pavement tiles were manufactured and assessed experimentally. The results reveal a promising potential of valorizing the local DM in the development of precast geopolymer products, despite the effects of shrinkage cracking on the experimental evaluation of the material mechanical properties.

Investigation of the Geopolymerization Potential of a Waste Silica-Rich Diabase Mud.

Diabase mud (DM) is a silica-rich residue yielding from aggregate crushing and washing operations in quarries. This work focuses on identifying the geopolymerization potential of a diabase mud through characterization of its mineralogical composition, investigation of its reactivity, and assessment of the early compressive strengths of alkali activated mixtures formulated based on the mud's dissolution results. The findings suggest that considerably low amounts of Al and Si metals were dissolved following the dissolution tests conducted on DM, however, the incorporation of small quantities of CEM I, gypsum, and metakaolin (MK) moderately at a Na2SiO3:NaOH ratio of 50:50 and with a molarity of NaOH of 4 M enhanced the geopolymerization compared to low L/S ratio mixtures cured at different conditions. When M was increasing, the high L/S ratio mixtures exhibited fluctuations in strengths, especially beyond a 10 M NaOH molarity. Maximum strengths of mixtures at equivalent molarity of 10 were achieved when the Na2SiO3:NaOH ratio reached 30:70, regardless of the ambient conditions and the presence of CEM I. The curing conditions, the ratio of Na2SO3:NaOH, and the presence of CEM I in the DM-based mixtures did not appear to significantly affect the mixture when NaOH concentration was between 2 M and 4 M; at higher molarities, however, these enhanced the strengths of the geopolymerized DM.

Effect of Mechanically Treated Recycled Aggregates on the Long Term Mechanical Properties and Durability of Concrete.

The objective of this research was to study the effect of an optimal mechanical treatment method to reduce the mortar adhered on recycled aggregates (RCA) on the long-term mechanical properties and durability of concretes containing RCA at different replacement levels. It was found that concretes incorporating treated RCA exhibited sharper and more significant increase on 90- and 365-day compressive strengths than any other investigated mixture. The same mixtures also benefitted from a 'shrinkage-controlling' effect, where strains and mass losses were reduced by almost 15% and 10%, respectively, compared to the reference concrete. While sulfate resistance and carbonation resistance are predominantly defined by the hydration products available within the cement paste and not to a large extent by the aggregate type and quality, the incorporation of either treated or untreated RCA in concrete did not appear to expose RACs to significant durability threats.

Optimizing the Mechanical Properties of Ultra-High-Performance Fibre-Reinforced Concrete to Increase Its Resistance to Projectile Impact.

This study describes an extensive experimental investigation of various mechanical properties of Ultra-High-Performance Fibre-Reinforced Concrete (UHPFRC). The scope is to achieve high strength and ductile behaviour, hence providing optimal resistance to projectile impact. Eight different mixtures were produced and tested, three mixtures of Ultra-High-Performance Concrete (UHPC) and five mixtures of UHPFRC, by changing the amount and length of the steel fibres, the quantity of the superplasticizer, and the water to binder (w/b) ratio. Full stress-strain curves from compression, direct tension, and flexural tests were obtained from one batch of each mixture to examine the influence of the above parameters on the mechanical properties. The Poisson's ratio and modulus of elasticity in compression and direct tension were measured. Additionally, a factor was determined to convert the cubic strength to cylindrical. Based on the test results, the mixture with high volume (6%) and a combination of two lengths of steel fibres (3% each), water to binder ratio of 0.16% and 6.1% of superplasticizer to binder ratio exhibited the highest strength and presented great deformability in the plastic region. A numerical simulation developed using ABAQUS was capable of capturing very well the experimental three-point bending response of the UHPFRC best-performed mixture.