The research, conducted by Hanieh Feyzi, Pouya Khosravi-Hajivand, and Mohammadjavad Hamidia of Shahid Beheshti University in Tehran, introduces a machine learning-aided methodology for identifying the seismic performance level of reinforced concrete moment-resisting frames. Rather than relying on fixed drift thresholds, the framework learns from what laboratory tests have actually shown about how connections behave. At its core is an experimental database containing force-deformation curves from 236 reinforced concrete beam-column connection subassemblies subjected to cyclic loading, meaning the specimens were pushed back and forth repeatedly in the way earthquake shaking alternately loads and unloads a structure. This database spans decades of testing campaigns from laboratories around the world, encompassing joints with widely varying detailing practices, from older non-ductile configurations to modern seismically detailed assemblies.

The team aligned their analysis with the performance framework of ASCE 41-23, the American Society of Civil Engineers standard for seismic evaluation and retrofit of existing buildings. That standard defines discrete performance levels, including Immediate Occupancy, at which a structure remains essentially functional with only minor damage; Damage Control, a intermediate state limiting degradation; Life Safety, at which significant damage occurs but the structure retains margin against partial collapse; and Collapse Prevention, the threshold beyond which the building is on the verge of losing vertical load-carrying capacity. The researchers extracted the peak drift ratios at which each of these performance levels was reached for every specimen in the database. What they found was striking: for specimens with diverse structural characteristics, the range of drift ratios associated with any single performance level was remarkably broad. A drift ratio that signals life-threatening damage in one connection might correspond to barely perceptible cracking in another.

This variability is precisely why the authors contend that the peak drift ratio alone cannot fully characterize the seismic performance of a structure. The finding has profound implications for engineering practice. Under current code-based linear design procedures, an engineer designs a frame, computes the projected peak drift ratio, and checks it against a single prescribed limit. If the number passes, the design is accepted. Yet the experimental record shows that the same projected drift can land a building in radically different damage states depending on the attributes of its beam-column connections, the critical regions where beams frame into columns and where seismic energy is dissipated. In effect, the code’s universal yardstick measures something real, but it measures it with a tolerance so wide that the actual performance outcome remains largely unknown until an earthquake reveals it.

To close that gap, the researchers turned to supervised classification. They trained and compared nine shallow machine learning models alongside one deep learning model, tasking each with mapping structural and geometric attributes of a connection, together with the projected peak drift ratio from code-based linear analysis, to the correct ASCE 41-23 performance level. The standout performer was the Extra Trees algorithm, an ensemble method that builds many randomized decision trees and aggregates their votes. On the held-out testing dataset, the Extra Trees classifier achieved 88 percent accuracy, a level of reliability that the authors argue is sufficient for practical design support. The deep learning model, despite its greater architectural complexity, did not outperform the shallow ensemble, a result consistent with a broader pattern in engineering applications where tabular data of moderate size favors tree-based learners over neural networks.

Robustness was a central concern of the methodology. Machine learning models in engineering risk a subtle failure mode: they can appear accurate while merely memorizing the quirks of their training data, a phenomenon known as overfitting. To guard against this, the team determined the optimal hyperparameters of each model through Bayesian optimization, implemented via the BayesSearchCV framework, which uses probabilistic surrogate models to search the hyperparameter space efficiently rather than exhaustively. They then validated generalizability using nested five-fold cross-validation, a rigorous scheme in which hyperparameter tuning and performance estimation are carried out on strictly separated data partitions. Nested cross-validation is widely regarded as one of the most trustworthy protocols for estimating how a model will behave on unseen data, and its use here signals a deliberate effort to meet the evidentiary standards of structural engineering rather than those of a machine learning benchmark competition.

Transparency posed another challenge. Black-box predictions carry little weight in a field where lives depend on the reasoning behind a design decision. To open the box, the researchers applied SHAP analysis, or SHapley Additive exPlanations, a technique borrowed from cooperative game theory that quantifies the contribution of each input feature to every individual prediction. The SHAP assessment revealed which structural and geometric attributes most strongly shaped the model’s judgments about performance levels, giving engineers a window into the learned relationships and a means of checking them against physical intuition. This interpretability layer matters not only for professional acceptance but also for scientific scrutiny, since it allows other researchers to verify that the model has learned plausible mechanics rather than statistical artifacts.

The framework’s practical value was demonstrated through a seismic design case study of a reinforced concrete moment frame. Following conventional code-prescribed linear procedures, the study showed how the projected peak drift ratio, fed into the trained classifier together with the relevant structural descriptors, yields a probabilistic identification of the resulting performance level. Instead of a binary pass-fail check against a fixed drift limit, the engineer receives a data-informed assessment of whether the design is likely to achieve Immediate Occupancy, Damage Control, Life Safety, or Collapse Prevention. This capability fits squarely within the broader movement toward performance-based seismic design, which seeks to give owners and engineers explicit control over the damage state a building will experience at given hazard levels, rather than merely ensuring that code minimums are met. Recent comparative studies have shown that performance-based approaches can produce materially different, and often more rational, designs than conventional force-based procedures.

The study also connects to a rapidly expanding research program on data-driven earthquake engineering. In recent years, machine learning has been applied to estimate drift capacity of reinforced concrete walls, classify earthquake damage to buildings from sensor and inspection data, predict seismic demands on steel moment frames, and even infer damage states from the visual features of surface crack patterns using computer vision. The present work extends this lineage in a distinctive direction: rather than assessing damage after an earthquake, it embeds learning models directly into the design workflow, using the outputs of standard linear analysis as inputs to a far more nuanced performance evaluation. The authors’ related research on post-earthquake performance identification using residual drift ratios suggests a complementary pathway, in which the same learning infrastructure could serve both pre-event design verification and post-event damage assessment.

The implications for practice are considerable. Codes such as ASCE 7 and ACI 318, along with Eurocode 8, continue to anchor seismic design in linear analysis with drift checks, and a wholesale replacement of these procedures is neither imminent nor necessarily desirable. But the study’s evidence suggests that supplementing the drift check with a learned performance classifier, calibrated on hundreds of laboratory tests and validated with rigorous statistical protocols, could give engineers a far sharper picture of what their designs will actually do when the ground shakes. The broad scatter of drift limits behind every code threshold means that many designs accepted today carry hidden uncertainty about their true performance level. A framework that reduces that uncertainty, while remaining interpretable and compatible with existing workflows, offers a pragmatic bridge between the codified simplicity of the past and the performance-based precision that the discipline has long aspired to. As experimental databases grow and learning models improve, the era in which a single number could stand in for a building’s seismic fate may finally be drawing to a close.

Subject of Research: Machine learning-aided performance-based seismic design of reinforced concrete moment-resisting frames

Article Title: Performance-based machine learning-aided seismic design methodology for reinforced concrete moment frames

Article References: Feyzi, H., Khosravi-Hajivand, P., & Hamidia, M. (2026). Performance-based machine learning-aided seismic design methodology for reinforced concrete moment frames. Bulletin of Earthquake Engineering. https://doi.org/10.1007/s10518-026-02598-3

Image Credits: AI Generated

DOI: 10.1007/s10518-026-02598-3

Keywords: seismic design, machine learning, reinforced concrete, moment-resisting frames, beam-column connections, ASCE 41-23, performance-based design, peak drift ratio, Extra Trees, Bayesian optimization, SHAP, earthquake engineering

The method applies to a class of buildings that dominates urban landscapes across the seismic world: regular reinforced concrete structures of two to eight stories, with story heights not exceeding four meters, and without significant structural irregularities. These are the apartment blocks, schools, offices and mixed-use buildings that line the streets of Istanbul, Athens, Santiago, Manila and countless other cities sitting near active faults. The procedure covers both frame-only systems, in which moment-resisting frames carry all lateral loads, and dual systems that combine frames with structural walls. By targeting precisely this building stock, the researchers have aimed their tool at the scale of the problem where improved preliminary design can save the most lives.

What makes the approach remarkable is its economy of input. Rather than demanding a complete structural model, the simplified method requires only estimated gravity loads, column tributary areas, the number of stories, story heights, plan areas, and the short-period design spectral acceleration for the site. From these basic quantities, the procedure uses fundamental mechanics of materials and minimum reinforcement requirements to proportion beams, columns and walls so that the resulting structure achieves adequate strength, stiffness and ductility at both the member level and the system level. In other words, the method embeds the core lessons of decades of earthquake engineering, capacity design principles and ductile detailing into a streamlined calculation sequence that a practicing engineer can complete in minutes rather than days.

The intellectual lineage of the procedure stretches back decades. Uğur Ersoy, one of Turkey’s most influential reinforced concrete researchers, developed the original formulation, which drew on international experience with simplified design guides, including the American Concrete Institute’s simplified design documents, Japanese seismic provisions, and the empirical wall-index observations of Shiga and colleagues after Japanese earthquakes. That heritage matters because the method is not an abstract exercise; it distills field evidence about which buildings survived past earthquakes and why. Buildings with sufficient walls, well-proportioned members and sensible proportions have repeatedly performed better in Chile, Japan and Turkey, and the new procedure translates those observations into explicit design rules for the modern code environment.

The real test of any simplified method is whether its output can survive scrutiny by the most demanding analysis tools available. The research team put 27 buildings, 11 frame-only structures and 16 dual frame-wall systems, through nonlinear static pushover analyses. Every one of these buildings was designed solely with the simplified method, using realistic and hypothetical configurations representative of actual practice. The pushover analyses pushed each structure laterally until its components reached defined performance limits, allowing the researchers to evaluate performance levels and shear demand-to-capacity ratios for every beam, column and wall according to ASCE 41-23, ACI 369.1-22 and ACI 318-25, three of the most authoritative standards in American structural engineering practice.

The results are striking. Under the design-basis earthquake, columns and walls in all of the evaluated buildings achieved at least the life safety performance level, the standard that ensures structural elements may be damaged but remain able to protect occupants. Beams, which are intentionally detailed to yield and dissipate energy before more critical members do, ranged between life safety and collapse prevention, exactly the hierarchy that modern capacity design seeks to enforce. Shear capacity proved adequate for nearly all members, with the only exceptions being a few short-span beams that fell outside the method’s stated applicability limits, a finding that reinforces the importance of respecting the boundaries within which any simplified tool is valid.

Comparisons with designs produced through the full TBER 2018 code route revealed another important property: the simplified method provides an acceptable level of conservatism. That conservatism manifests primarily in member dimensions rather than in reinforcement quantities, meaning the procedure tends to produce somewhat larger columns, beams and walls while keeping steel demands reasonable. For practitioners, this is a reassuring outcome. It suggests the method erringly errs on the side of robustness, giving early-stage designs a comfortable margin that will be refined, but not radically overhauled, in the final code-compliant design phase that follows.

The urgency behind this work is impossible to separate from recent seismic tragedy. The February 2023 Kahramanmaraş earthquake doublets in Türkiye, with magnitudes of 7.8 and 7.6, caused catastrophic damage to thousands of reinforced concrete buildings and prompted a wave of reconnaissance studies documenting failures that better preliminary design could have prevented. The study’s reference list is punctuated with field reports from that disaster and from earlier Turkish earthquakes in Kocaeli, Bingöl and Van, alongside lessons from the 1985 Chile and 1995 Kobe events. This context gives the research its emotional weight: every paragraph of the validation effort is ultimately about preventing the next collapse, particularly in a country where millions of people live in buildings of exactly the type the method addresses.

The researchers are careful to position the tool correctly within the design workflow. The simplified method is a preliminary design procedure, not a substitute for final structural design. It dispenses with elaborate analysis at the concept stage, when architects and engineers are negotiating layouts, column grids and wall locations, and it guarantees instead a minimum acceptable level of seismic performance from the very first sketch. The final code-compliant design, with its full three-dimensional modeling, modal response spectrum or time-history analysis, and detailed reinforcement design, is performed subsequently. By front-loading seismic competence into the earliest decisions, the method prevents the most dangerous failure mode of the design process itself: the discovery, late in the project, that the chosen architecture cannot be made seismically sound without painful and expensive redesign.

The authors conclude that the simplified method offers a practical, intuitive, rapid and reliable tool for the preliminary seismic design of regular low- to medium-rise reinforced concrete buildings. In an era when performance-based design has become increasingly sophisticated and computationally demanding, there is something quietly revolutionary about a procedure that argues for simplicity as a safety feature. Complex codes are only as effective as the practitioners who apply them correctly, and the gap between cutting-edge analysis and everyday practice is where many seismic vulnerabilities are born. By closing that gap with a method grounded in mechanics, validated against modern performance standards, and now written into a national seismic code, this research demonstrates that sometimes the most powerful engineering innovation is the one that makes doing the right thing the easiest thing to do. For the residents of eight-story apartment buildings from Istanbul to wherever the next great earthquake will strike, that simplicity may prove to be the difference between a shaken building and a collapsed one.

Subject of Research: Simplified preliminary seismic design of low-to-medium rise reinforced concrete buildings

Article Title: A simple preliminary seismic design procedure for low-to-medium rise reinforced concrete buildings

Article References: A simple preliminary seismic design procedure for low-to-medium rise reinforced concrete buildings. (n.d.). https://doi.org/10.1007/s10518-026-02696-2

Image Credits: AI Generated

DOI: 10.1007/s10518-026-02696-2

Keywords: seismic design, reinforced concrete, earthquake engineering, preliminary design, TBER 2018, ASCE 41-23, pushover analysis, dual systems, moment-resisting frames, life safety, capacity design, Bulletin of Earthquake Engineering