Dr. Lenganji Simwanda, PhD., R.Eng, P.EIZ

Researcher in Computational Civil and Structural Engineering

Dr. Lenganji Simwanda is a researcher in Civil and Environmental Engineering, specializing in computational methods. His work integrates advanced technologies such as artificial intelligence (AI), machine learning (ML), structural reliability, and uncertainty quantification to tackle critical challenges in civil infrastructure systems. His primary focus is on enhancing the structural integrity and sustainability of ultra-high-performance concrete (UHPC) and industrial infrastructure through probabilistic modeling and data-driven approaches.

Dr. Simwanda's research extends beyond concrete infrastructure, encompassing collaborations with experts in light-gauge steel structures, particularly cold-formed steel, and chemists working on sustainable energy storage solutions. Together, they explore innovative materials such as pillared interlayer alumina-clays and geopolymers, with a focus on their applications in hydrogen storage and sustainability.

Currently, Dr. Simwanda is exploring the use of generative AI models, including diffusion models, to optimize structural designs and material compositions. His research aims to overcome the limitations posed by small datasets, ensuring efficient and transparent AI applications in engineering. Inspired by frameworks like RFdiffusion, he is investigating how generative models can drive the discovery of new materials or structural innovations in civil engineering.

Get In Touch With Dr. Simwanda

Reach out for collaborations, inquiries, or to share insights.

About

Postdoctoral Fellow @ Czech Technical University in Prague

Part-Time Structural Designer @ V-CON s.r.o. , Valbek Group

Education

2022-12 Ph.D. Civil Engineering, Stellenbosch University, South Africa

2021-05 M.Eng. Structural Engineering, University of Zambia, Zambia

2019-03 B.Eng. Structural Engineering, University of Zambia, Zambia

Research Interests

Civil and Structural Engineering | Sustainability | Artificial Intelligence | Risk, Reliability, and Resilience

Selected Experience

2025-01 Partime Structural Designer, V-CON s.r.o Valbek Group,, Czech Republic [Current Position]

2024-02 Postdoctoral Fellow, Klokner Institute, Czech Technical University in Prague, Czech Republic [Current Position]

2023-07 Postdoctoral Fellow, University of South Africa, South Africa

2023-01 Consolidoc Postdoctoral Fellow, Stellenbosch University, South Africa

Publications

2025

Epistemic uncertainties in torque capacity prediction models for circular CFDST members

Authors Lenganji Simwanda, B.D. Ikotun, F.M. Ilunga, E.K. Onyari

Journal Journal of Constructional Steel Research

Volume 226

Pages 109299

View Abstract View article at publisher

List of Journal Publications

2025

Machine learning prediction of web crippling strength in cold-formed steel beams with

staggered slotted perforations

Authors Perampalam Gatheeshgar, R.S.S. Ranasinghe, Lenganji Simwanda, D.P.P. Meddage, Damith Mohotti

Journal Structures

Volume 71

Pages 108079

View Abstract View article at publisher

2024

Suitability of Mechanics-Based and Optimized Machine Learning-Based Models in the

Shear Strength Prediction of Slender Beams Without Stirrups

Authors Abayomi B. David, Oladimeji B. Olalusi, Paul O. Awoyera, Lenganji Simwanda

Journal Buildings

Volume 14

Pages 3946

View Abstract View article at publisher

2024

Explainable machine learning models for predicting the ultimate bending capacity of

slotted perforated cold-formed steel beams under distortional buckling

Authors Lenganji Simwanda, P, Gatheeshgar, FM Ilunga, BD. Ikotun, SM. Mojtabaei, EK. Onyari

Journal Thin-Walled Structures

Volume 205 part C

Pages 112587

View Abstract View article at publisher

2024

Multilinear regression model for predicting the ultimate load of slender circular CFDST

columns subjected to concentric and eccentric loading

Authors Akuta Usongo Amika, Trevor N. Hass, Lenganji Simwanda

Journal Structural Concrete

Volume 25

Pages n/a - n/a

View Abstract View article at publisher

2024

Reliability analysis of shear design provisions for cold formed steel sections

Authors Lenganji Simwanda, P. Gatheeshgar, BD. Ikotun, M.Bock, EK. Onyari, , FM Ilunga

Journal Journal of Constructional Steel Research

Volume 217

Pages 108656

View Abstract View article at publisher

2024

Prediction of Torque Capacity in Circular Concrete-Filled Double-Skin Tubular Members

under Pure Torsion via Machine Learning and Shapley Additive Explanations Interpretation

Authors Lenganji Simwanda, Bolanle Deborah Ikotun

Journal Buildings

Volume 14

Pages 1040

View Abstract View article at publisher

2024

Numerical Study of Thermal Efficiency in Light-Gauge Steel Panels Designed with

Varying Insulation Ratios

Authors Dilanka Chandrasiri, Perampalam Gatheeshgar, Hadi Monsef Ahmadi, Lenganji Simwanda

Journal Buildings

Volume 14

Pages 300

View Abstract View article at publisher

2023

Bayesian calibration and reliability analysis of ultra high-performance fibre reinforced

concrete beams exposed to fire

Authors Lenganji Simwanda, Adewumi John Babafemi, Nico De Koker, Celeste Viljoen

Journal Structural Safety

Volume 103

Pages 102352

View Abstract View article at publisher

2023

Numerical modelling, parametric analysis and design of UHPFRC beams exposed to fire

Authors Lenganji Simwanda, Charles Kahanji, Faris Ali

Journal Structures

Volume 52

Pages 1-16

View Abstract View article at publisher

2023

Structural reliability of ultra high-performance fibre reinforced concrete beams in shear

Authors Lenganji Simwanda, Nico de Koker, Celeste Viljoen, Adewumin John Babafemi

Journal Structural Concrete

Volume 24

Pages 2862-2878

View Abstract View article at publisher

2023

Structural reliability of ultra high-performance fibre reinforced concrete beams in flexure

Authors Lenganji Simwanda, Nico de Koker, Celeste Viljoen

Journal Engineering Structures

Volume 244

Pages 112767

View Abstract View article at publisher

Presentations

Ultra-High-Performance Concrete: Probabilistic models, Model Uncertainty, and Artificial Intelligence

Abstract

The presentation explores the use of probabilistic models and AI to enhance Ultra-High-Performance Concrete (UHPC). It discusses improving UHPC mix designs with AI and addresses uncertainties in resistance models. Key topics include the underestimated tensile strength contribution in traditional models and AI-driven optimization for performance and sustainability. Ongoing research and collaborative efforts to advance UHPC applications in structural engineering are highlighted

Date:    21 November 2024   

Venue: Klokner Institute- Czech Technical University in Prague

Probabilistic Models for UHPC in Bending: A State-of-the-Art Perspective.

Date:    2 October 2024   

Venue: Politecnico di Torino

Abstract

The presentation discusses probabilistic models for Ultra-High Performance Concrete (UHPC) in bending applications. Addressing the lack of mature design guidelines for UHPC, the presentation highlights the uncertainties in current models at material, geometric, and model levels. It proposes advancements in probabilistic models to refine these guidelines, emphasizing the need for large datasets and detailed assessments to improve UHPC's accuracy and reliability in structural applications. The talk outlines future directions for enhancing model recalibration and managing inherent uncertainties.

Model Uncertainty in European UHPC Standards: Insights from SIA-2052 and NF P18-710 Flexure Models

Date:    23 September 2024   

Venue: Hotel Galant - Mikulov

Abstract

The presentation addresses model uncertainty in the design of Ultra-High-Performance Concrete (UHPC) beams within European standards. It highlights the discrepancies and limitations in the current models (Swiss SIA 2052 and French NF P18-710) through a robust analysis involving 211 experimental beam data points. This study assesses the bias and variability inherent in these models, and explores the implications for partial factor method verifications in design standards. Recommendations are made for refining these models using non-linear finite element analysis to improve design accuracy and efficiency

View Publication On ResearchGate