The research, conducted by Ghazi Bahroz Jumaa of the Civil Engineering Department at the University of Garmian in the Kurdistan Region of Iraq, tackles one of the most safety-critical questions in reinforced concrete design: how much shear force can a beam resist before it fails catastrophically? Shear failure is notoriously brittle and sudden, unlike the gradual, warning-giving flexural failure that designers prefer to govern. For beams without stirrups, the closed steel hoops that normally provide shear reinforcement, the entire resistance must come from the concrete itself, making accurate prediction of that resistance a matter of structural life and death.

The heart of the problem lies in the messy, heterogeneous nature of RCA. When old concrete is crushed, each recycled particle is a composite: natural stone wrapped in a layer of adhered mortar that is often weaker, more porous, and riddled with micro-cracks from the crushing process. In a loaded beam, diagonal cracks propagate through the concrete web, and the rough crack faces resist sliding through a mechanism engineers call aggregate interlock. Weaker, more irregular recycled particles interlock less effectively, and the old mortar attached to them introduces additional weak zones. The result is a material whose shear behavior is inherently more variable than that of natural-aggregate concrete, and whose performance depends on the percentage of recycled content, a variable that most established design equations were never formulated to include.

Existing design provisions, such as those in the American Concrete Institute’s ACI 318 code and Europe’s Eurocode 2, were calibrated decades ago on natural-aggregate concrete. When applied to RCA beams, they can be either unconservative or wastefully conservative, and the experimental literature over the past two decades, from studies by Han and colleagues in 2001 through work by Rahal and Alrefaei, Ignjatović and co-workers, and many others, has documented a wide scatter in measured shear strengths. Empirical equations such as the classic Zsutty formulation, dating back to 1971, capture broad trends but struggle with the added complexity of recycled content. This is precisely the kind of nonlinear, multi-variable prediction problem where machine learning has begun to shine.

Jumaa assembled a database of 179 experimental RCA beam specimens drawn from published laboratory studies spanning more than two decades. Each specimen was characterized by the key parameters that govern shear capacity: beam width, effective depth, the shear span-to-depth ratio, the longitudinal reinforcement ratio, the concrete compressive strength, and, crucially, the percentage of recycled concrete aggregate replacing natural stone. The measured shear capacities ranged from small laboratory beams failing at around 12 kilonewtons to massive 400-millimetre-wide specimens carrying more than 860 kilonewtons, giving the models a rich and demanding training landscape.

Five machine learning algorithms were trained and compared under a rigorous validation protocol: k-Nearest Neighbors, Random Forest, Support Vector Machines, Gaussian Process Regression, and Artificial Neural Networks. Rather than relying on a single train-test split, which can flatter a model by luck of the draw, the study employed a 10-fold cross-validation scheme repeated 20 times. In this procedure the data are repeatedly shuffled and divided into ten subsets; the model trains on nine and is tested on the held-out tenth, with the process cycled through all folds and then repeated from scratch twenty times to average out randomness. This is the gold standard for demonstrating that a model’s accuracy reflects genuine learning of underlying physics rather than memorization of individual specimens.

The results were striking. Support Vector Machines emerged as the top performer, achieving a coefficient of determination of 0.933, meaning the model explained more than 93 percent of the variance in measured shear strength, with a root mean square error of 29.98 kilonewtons and a mean absolute error of just 13.08 kilonewtons. Gaussian Process Regression followed closely with an R-squared of 0.912 and the lowest mean absolute error of the trio at 11.81 kilonewtons, while Artificial Neural Networks achieved an R-squared of 0.892. All three outperformed both the newly proposed empirical equations, including a modified Zsutty equation and an ACI 318-style equation incorporating RCA content, and the traditional design codes themselves. Support Vector Machines, which work by mapping inputs into a high-dimensional feature space where a maximum-margin boundary separates the data, proved especially adept at handling the nonlinear interactions between beam geometry, material strength, and recycled content.

Beyond raw accuracy, the study included a parametric analysis that interrogated how the trained models respond when individual variables change, a crucial sanity check for any data-driven engineering tool. The machine learning predictions reproduced physically sensible trends: shear capacity increased with beam depth and width, and decreased as the recycled aggregate content rose, consistent with the known degradation of aggregate interlock in RCA concrete. This alignment between statistical learning and structural mechanics matters enormously, because a black-box model that contradicts physical intuition would be dangerous to trust, no matter how impressive its error metrics appear on paper.

The practical implications reach well beyond the laboratory. As governments worldwide push circular-economy mandates and green building standards, the volume of RCA entering structural applications is set to grow sharply, and the fib Model Code 2020 and the next generation of Eurocode 2 are actively working toward codified design rules for recycled aggregate concrete structures. Reliable predictive tools are the prerequisite for such codification: engineers need confidence that a beam designed with, say, 50 or 100 percent recycled coarse aggregate will carry its intended shear load with the same margin of safety as a conventional beam. A validated machine learning model, trained on nearly two hundred real experiments, offers exactly that confidence, and could inform both code committees drafting new provisions and practitioners evaluating existing structures built with sustainable materials.

The study also fits into a broader movement sweeping through structural engineering, in which data-driven methods complement, and sometimes surpass, hand-derived empirical formulas. Researchers have already applied neural networks and related techniques to predict the shear capacity of fiber-reinforced polymer concrete beams, the compressive strength of lightweight and recycled concretes, and bond strength in reinforced concrete members. What distinguishes the present work is its focus on the specific, under-addressed case of stirrup-free RCA beams and its head-to-head comparison of five algorithms under a demanding cross-validation protocol. As Jumaa notes in the paper, the work represents a step forward in using machine learning to enhance structural engineering, with the potential to make a significant impact in the field of sustainable materials. If the construction industry is to recycle its own waste at scale, it may well need algorithms, as much as new mix designs, to make the resulting structures safe.

Subject of Research: Machine learning prediction of shear strength in reinforced recycled aggregate concrete beams without stirrups

Article Title: Shear strength prediction of reinforced recycled aggregate concrete beams without stirrups using soft computing and empirical methods

Article References: Jumaa, G. B. (2026). Shear strength prediction of reinforced recycled aggregate concrete beams without stirrups using soft computing and empirical methods. Neural Computing and Applications, 38(19), Article 771. https://doi.org/10.1007/s00521-026-12488-z

Image Credits: AI Generated

DOI: 10.1007/s00521-026-12488-z

Keywords: machine learning, recycled concrete aggregate, shear strength, reinforced concrete beams, stirrups, Support Vector Machines, artificial neural networks, Gaussian Process Regression, ACI 318, Eurocode 2, sustainable construction, structural engineering

The idea behind recycled-component concrete is deceptively simple. Instead of crushing waste concrete down into small recycled aggregates, or even into mid-sized concrete lumps of 60 to 300 millimeters, the new approach salvages entire concrete members, such as columns recovered through selective dismantling, and gives them only minimal processing: stripping off the rendering layer, cutting to length, and roughening the surfaces. These chunky recycled components are then placed directly inside the formwork of a new member, and fresh self-compacting concrete is cast around them. The result is a hybrid column in which nearly half the cross-section is made of old, already-cured concrete, complete with its original reinforcing bars and stirrups still intact inside.

The advantages over conventional recycling are substantial. Crushing waste into recycled aggregates still demands all the cement of ordinary concrete, so the carbon penalty of clinker production remains untouched. Recycled lump concrete improves on that by replacing some of the new mix, but casting quality limits lump replacement to about 35 percent, and the crushing process generates fine powder that is difficult to use. Recycled-component concrete, by contrast, can replace more than 50 percent of the new concrete, produces far less waste powder, and preserves the old reinforcement in usable condition, potentially trimming the amount of new steel required. Earlier studies in Europe had reused cut concrete slabs directly in new buildings, with one prototype floor system showing embodied carbon 80 to 94 percent lower than conventional concrete floors. But direct reuse runs into a stubborn problem: salvaged members rarely match the dimensions or strength grades their new positions demand. Casting new concrete around a recycled component elegantly dissolves that mismatch, because the surrounding mix absorbs any dimensional discrepancy and a stronger new mix can compensate for weaker old concrete.

What had never been tested, until now, was whether this hybrid configuration could survive the two brutal trials that matter most for columns: crushing axial compression and repeated cyclic earthquake loading. The concern was real. In a recycled-component concrete column, the old concrete sits as a concentrated block in the center of the section, surrounded by new concrete, creating a large, regularly shaped interface between the two generations of material. If that interface were to delaminate or slip under load, the whole concept would collapse, figuratively and literally. There was also a puzzle for the engineers’ calculators: the column contains two overlapping systems of stirrups, the new outer hoops and the old hoops locked inside the recycled component, and existing strength-prediction models, built for a single confinement system, simply did not apply.

To find out, the team fabricated nine square column specimens, 400 millimeters on a side and 1200 millimeters tall, for axial compression testing, seven of them recycled-component columns with a replacement ratio of 46 percent and two conventional references. Each recycled component was a 270 by 270 millimeter column, cut and roughened from a prefabricated 300 by 300 millimeter old column, its eight 12-millimeter longitudinal bars and 6-millimeter stirrups retained. The specimens were squeezed in a 15,000-kilonewton machine under displacement control, with strain gauges glued to both old and new stirrups to reveal how the two confinement systems shared the work. Seven more specimens, cantilever columns with a shear-span ratio of 3.375, were subjected to low-cycle reversed lateral loading under a constant axial load ratio of 0.4, mimicking the demands of a major earthquake.

The axial results were encouraging. During testing, no large-area delamination or spalling of new concrete from the recycled components occurred, confirming that the bond between old and new material held firm all the way to failure. The recycled-component columns carried somewhat lower peak loads than the references, from 7571 to 8339 kilonewtons against 8445 to 8648 kilonewtons, but the shortfall traced cleanly to the weaker old concrete, not to any interfacial weakness. Crucially, when the team raised the strength of the new concrete, specimen CR6 reached 8339 kilonewtons, nearly matching the 8445-kilonewton conventional reference, proving that a stronger surrounding mix can compensate for weaker salvaged material. Inclined-crossed stirrups, arranged in an X-pattern instead of conventional rectangular hoops, slightly increased capacity at the same steel quantity, and strain measurements showed they mobilized their confinement more effectively.

The seismic tests delivered the headline finding. Under reversed cyclic loading, the recycled-component columns and the conventional reference failed in the same characteristic way, with flexural cracks forming at the base, evolving into diagonal flexural-shear cracks, and culminating in concrete crushing within the plastic hinge region. Their hysteretic loops were similarly full, indicating strong energy dissipation, and their skeleton curves were nearly indistinguishable. Ductility coefficients ranged from 3.43 to 3.80, comfortably above the minimum of 3 required for reinforced concrete columns, and all but one specimen exceeded the 2 percent ultimate drift ratio demanded by Chinese seismic design codes. Remarkably, the recycled-component columns actually developed higher peak lateral loads than the conventional reference, reaching up to 370.2 kilonewtons against 343.9 kilonewtons, thanks to the stronger old concrete and the higher yield strength of the retained reinforcement. Incorporating a recycled component while slimming the new longitudinal bars from 18 and 16 millimeters down to 12 millimeters cost only 4 percent of ductility.

The experiments also delivered practical verdicts on construction details. Drilling post-installed rebars into the recycled components proved counterproductive under axial load, reducing peak capacity by 4.2 percent, apparently because the drilling introduced internal defects, and offering no seismic benefit either. Welded connectors between new and old stirrups improved ductility by only 4.7 percent, a gain judged too small to justify the cumbersome installation. The team’s recommendation is refreshingly economical: skip both. On the other hand, widening the spacing of the new stirrups cut the equivalent viscous damping coefficient by 17.6 percent and dropped one specimen below the drift-ratio requirement, so the confinement steel in the plastic hinge zone must not be skimped. Inclined-crossed stirrups emerged as the preferred arrangement, boosting energy dissipation by up to 8.6 percent, and their orientation relative to the loading direction made virtually no difference, freeing contractors from fussy placement.

Finally, the researchers closed the analytical gap. They divided the hybrid cross-section into three regions: unconfined cover concrete, concrete confined by the new stirrups alone, and the central old concrete squeezed simultaneously by both stirrup systems, then summed the regional contributions along with the old and new longitudinal steel. The resulting formula predicted the measured axial capacities of all nine specimens with a maximum error of just 8 percent, and only 5 percent for the conventional columns, showing that the dual-confinement model works across the spectrum. Taken together, the study offers the first technical foundation for columns that swallow nearly half their volume in salvaged concrete and part of their steel in salvaged reinforcement, while standing up to earthquakes as reliably as their all-new counterparts. In a construction industry desperate to cut its carbon ledger, that is a result worth building on.

Subject of Research: Axial compressive and seismic performance of recycled-component concrete columns made from reused waste concrete members

Article Title: Axial compressive behavior and seismic performance of recycled-component concrete columns

Article References: Li, Y., Wu, B., Zhang, H., & Wang, F. (2026). Axial compressive behavior and seismic performance of recycled-component concrete columns. Case Studies in Construction Materials, 25, Article e06453. https://doi.org/10.1016/j.cscm.2026.e06453

Image Credits: AI Generated

DOI: 10.1016/j.cscm.2026.e06453

Keywords: recycled-component concrete, recycled concrete, construction and demolition waste, seismic performance, axial compression, confinement, stirrups, carbon reduction, sustainable construction, cyclic loading, structural engineering, circular economy