Explainable machine learning using real, synthetic and augmented fire tests to predict fire resistance and spalling of RC columns

Naser, M.Z.; Kodur, Venkatesh

doi:10.1016/j.engstruct.2021.113824

Cited by 44 publications

(13 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A PDP displays the marginal effect of an individual feature while holding other features constant on model predictions [ 43 ]. A more thorough discussion on the fundamentals of explainability measures can be found elsewhere [ 42 , 43 ] as well as in our recent work [ 37 ], which is tailored for engineering applications, and [ 35 ], which tackled the fire response of concrete columns.…”

Section: Data Collection and Methodologymentioning

confidence: 99%

See 1 more Smart Citation

Discovering Graphical Heuristics on Fire-Induced Spalling of Concrete Through Explainable Artificial Intelligence

Tapeh

Naser

2022

Fire Technol

Self Cite

View full text Add to dashboard Cite

Fire-induced spalling of concrete continues to be an intriguing and intricate research problem. A deep dive into the open literature highlights the alarming discrepancy and inconsistency of existing theories, as well as the complexity of predicting spalling. This brings new challenges to creating fire-safe concretes and primes an opportunity to explore modern methods of investigation to tackle the spalling phenomenon. Thus, this paper leverages the latest advancements in explainable Artificial Intelligence (XAI) to vet existing theories on fire-induced spalling and to discover solutions/heuristics to predict spalling of concrete mixtures. The developed heuristics are in the form of graphs and nomograms . The proposed solutions allow interested researchers and engineers to graphically identify the propensity of a given concrete mixture to spalling directly and with ease. For example, we report that concrete mixtures with a combination of moderate water/binder ratio (of about 0.3), low heating rate (less than 2.5°C/min), moderate rise in temperature (less than 500°C), and have moisture content (less than 3%) are expected to be less prone to spalling. Further, findings from this research showcase the potential for developing simple (i.e., one-shot) and graphical (coding-free and formula-free) XAI-based solutions and web applications to address decades-long problems in the area of concrete research.

show abstract

Section: Data Collection and Methodologymentioning

confidence: 99%

“…Much of the cited works on the AI front, except [ 35 ], applied blackbox AI. In such AI, the logic and rationale behind models’ predictions are opaque.…”

Section: Introductionmentioning

confidence: 99%

Discovering Graphical Heuristics on Fire-Induced Spalling of Concrete Through Explainable Artificial Intelligence

Tapeh

Naser

2022

Fire Technol

Self Cite

View full text Add to dashboard Cite

show abstract

“…Perhaps modelers can use a variety of material models (to describe property degradation or mix real with synthetic/augmented data, which may minimize the noted skewness [30]. However, one must remember that replicating randomness effects may not be as easily obtained-especially when high nonlinearity or instability exists.…”

Section: Space Of Possible Featuresmentioning

confidence: 99%

Demystifying Ten Big Ideas and Rules Every Fire Scientist & Engineer Should Know About Blackbox, Whitebox and Causal Artificial Intelligence

Naser

2022

Fire Technol

View full text Add to dashboard Cite

Artificial intelligence (AI) is paving the way towards Fire 4.0, and the next few years will be elemental to how this technology transforms our academia, practice, and entrepreneurship. Despite the growing interest between fire researchers and practitioners, AI remains absent from our curriculum and we continue to lack a methodical framework to adopt, apply and create AI solutions suitable for our domain. To grow beyond the traditionally favored three-stepped linear process to AI (i.e., collect data fi apply algorithms fi arrive at predictions) and in pursuit of negating the notion of ''history repeats itself'' (e.g., look at the continued debate with regard to modernizing standardized fire testing, etc.), it is the motivation behind this letter to jump-start strategic discussions on the front of AI & Fire.In parallel, this letter intends to demystify some of the most fundamental concepts and to clear common misconceptions specific to AI on the frontier of fire. Broadly, this letter comprises two sections. The first section outlines big ideas pertaining to AI, and answers some of the burning questions with regard to the merit of adopting AI in our domain. The second section presents a set of rules 1 or technical recommendations an AI user may deem helpful to practice whenever AI is used as a method of investigation. The presented set of rules is complementary to the big ideas. One can think of this short letter as a companion to the Smart Systems in Fire Engineering special issue sponsored by Fire Technology [1]-where an

show abstract

“…Studies such as concrete analysis under extreme load conditions, corrosion risk estimations, methods for predicting resistance in concrete elements under the action of fire, and estimation of durability on concrete specimens have been performed with computational approaches Naser (2021a); Guzmán-Torres et al (2021b); Naser and Kodur (2022); Naser (2021b); Guzmán-Torres (2022).…”

Section: Introductionmentioning

confidence: 99%

Damage detection on steel-reinforced concrete produced by corrosion via YOLOv3: A detailed guide

Guzmán-Torres

Domínguez-Mota

Martínez-Molina

et al. 2023

Front. Built Environ.

View full text Add to dashboard Cite

Damage assessment applied to reinforced concrete elements is one of the main activities of infrastructure maintenance tasks. Among these elements, the problem of corrosion in reinforced concrete is particularly critical and requires careful consideration. Annually, governments invest a large amount of economic resources in this activity. However, most methodologies for damage assessment rely on visual inspection, which may be subjectively interpreted, producing inconsistent results and requiring a considerable amount of time and resources. This study evaluates the performance of real-time object detection using You Only Look Once, version 3, for detecting corrosion damage in concrete structures. The architecture of YOLOv3 is based on a complex, but efficient, convolutional neural network fed by a dataset proposed and labeled by the authors. Two training stages were established to improve the model precision, using transfer learning with medium- and high-resolution training images. The test results show satisfactory concrete-corrosion detection through validation photographs and videos demonstrating the capabilities of explainable artificial intelligence and its applications in civil engineering.

show abstract

Explainable machine learning using real, synthetic and augmented fire tests to predict fire resistance and spalling of RC columns

Cited by 44 publications

References 51 publications

Discovering Graphical Heuristics on Fire-Induced Spalling of Concrete Through Explainable Artificial Intelligence

Discovering Graphical Heuristics on Fire-Induced Spalling of Concrete Through Explainable Artificial Intelligence

Demystifying Ten Big Ideas and Rules Every Fire Scientist & Engineer Should Know About Blackbox, Whitebox and Causal Artificial Intelligence

Damage detection on steel-reinforced concrete produced by corrosion via YOLOv3: A detailed guide

Contact Info

Product

Resources

About