Gustavo G. Souza
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Gustavo G. Souza has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Andre Tavares
Engineering Department, Charleston Southern University
4 shared publications - Claudio Antonio Federico
Instituto de Estudos Avançados
4 shared publications - Edson Ramos Andrade
Military Institute of Engineering
4 shared publications - Raquel Aparecida A. Costa e Oliveira
Military Institute of Engineering
3 shared publications - Rodrigo Carneiro Curzio
Nuclear Engineering Graduate Program, Military Institute of Engineering (IME), Rio de Janeiro, Brazil
3 shared publications - Antônio César S. Leite
Nuclear Engineering Graduate Program, Military Institute of Engineering (IME), Brazil
2 shared publications
Latest Publications
- Open Access
- 14 October 2026
- Published by Elsevier in SSRN Electronic Journal
Integrated Framework for Decision-Making in Radiological Emergencies: A CNN–LSTM-Attention, CVaR, and Many-Objective Optimization Approach
Decision support in radiological emergencies requires dispersion forecasts, tail-risk dose estimates, and evacuation plans that balance conflicting goals. We present a framework that couples a convolutional and recurrent neural network with temporal attention (CNN–LSTM), used as a fast surrogate of a Gaussian plume model; a Conditional Value-at-Risk (CVaR) dose objective; five-objective evacuation optimization with the NSGA-III evolutionary algorithm; and four radiological indices. Each component is tested against an explicit reference. Trained on 40 years of ERA5 reanalysis (1985–2024), the surrogate achieved R² = 0.855 on a temporally independent test set (2019–2024) and, in a pre-registered test, reduced the mean absolute error relative to persistence by 17.2% (95% block-bootstrap confidence interval 15.6–18.7%). Against the Project Prairie Grass tracer measurements, the reference plume reproduced the mean concentration (geometric mean bias 1.04) but with large case-to-case scatter (fraction within a factor of two 0.475), and did not meet the pre-registered acceptance criterion. On a synthetic ten-zone planning area with objectives for evacuation time, dose CVaR, cost, logistic access, and collective panic, a pre-registered comparison over 30 independent runs ranked NSGA-III ahead of NSGA-II and a decomposition-based algorithm on two standard quality indicators (Wilcoxon p < 0.01), with a small Hypervolume margin over NSGA-II (0.8%). Under a single declared specification of 10,000 synthetic scenarios, the ranking of scenarios by the cognitive-bias coefficient was robust to its assumed weights (Spearman ρ ≥ 0.90). Motivated by the 1987 Goiânia accident, the framework is released with its code, pre-registrations, and an automatic check of every reported number against the analysis outputs.
- 14 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessIntegrated Framework for Decision-Making in Radiological Emergencies: A CNN–LSTM-Attention, CVaR, and Many-Objective Optimization Approach
Decision support in radiological emergencies requires dispersion forecasts, tail-risk dose estimates, and evacuation plans that balance conflicting goals. We present a framework that couples a convolutional and recurrent neural network with temporal attention (CNN–LSTM), used as a fast surrogate of a Gaussian plume model; a Conditional Value-at-Risk (CVaR) dose objective; five-objective evacuation optimization with the NSGA-III evolutionary algorithm; and four radiological indices. Each component is tested against an explicit reference. Trained on 40 years of ERA5 reanalysis (1985–2024), the surrogate achieved R² = 0.855 on a temporally independent test set (2019–2024) and, in a pre-registered test, reduced the mean absolute error relative to persistence by 17.2% (95% block-bootstrap confidence interval 15.6–18.7%). Against the Project Prairie Grass tracer measurements, the reference plume reproduced the mean concentration (geometric mean bias 1.04) but with large case-to-case scatter (fraction within a factor of two 0.475), and did not meet the pre-registered acceptance criterion. On a synthetic ten-zone planning area with objectives for evacuation time, dose CVaR, cost, logistic access, and collective panic, a pre-registered comparison over 30 independent runs ranked NSGA-III ahead of NSGA-II and a decomposition-based algorithm on two standard quality indicators (Wilcoxon p < 0.01), with a small Hypervolume margin over NSGA-II (0.8%). Under a single declared specification of 10,000 synthetic scenarios, the ranking of scenarios by the cognitive-bias coefficient was robust to its assumed weights (Spearman ρ ≥ 0.90). Motivated by the 1987 Goiânia accident, the framework is released with its code, pre-registrations, and an automatic check of every reported number against the analysis outputs.
- 20 November 2025
- Published by SAGE Publications in The Journal of Defense Modeling and Simulation: Applications, Methodology, Technology
Combat engineering risk assessment in radiologically overlapped zones under tactical operations
This study evaluates how radioactive contamination impacts a hypothetical human population in a military combat engineering context involving navigating radiologically contaminated environments. The model considers key risk metrics, including estimated radiation exposure levels and the potential development of radiation-induced leukemia, while accounting for age and sex as important variables in radiological risk assessment. Using analytical simulations with HotSpot Health Physics software, estimates were made for individuals continuously within the 700 mSv isodose zone, a dose also linked to immediate deterministic effects such as acute radiation syndrome. To improve operational efficiency and resource use in Chemical, Biological, Radiological, and Nuclear scenarios, the study discusses recommended engineering countermeasures, including optimizing route planning and protective barriers. It introduces a new proposed risk factor model that aims to measure the effects of overlapping radioactive plumes caused by changing local wind directions, highlighting increased vulnerability in transition zones. Unlike previous military or civilian radiological risk studies, this research focuses on the implications of plume interactions in real-time tactical environments, offering a useful framework for enhancing the safety and effectiveness of combat engineering operations facing radiological threats in hostile settings.
- 20 November 2025
- Published by SAGE Publications in The Journal of Defense Modeling and Simulation: Applications, Methodology, Technology
Combat engineering risk assessment in radiologically overlapped zones under tactical operations
This study evaluates how radioactive contamination impacts a hypothetical human population in a military combat engineering context involving navigating radiologically contaminated environments. The model considers key risk metrics, including estimated radiation exposure levels and the potential development of radiation-induced leukemia, while accounting for age and sex as important variables in radiological risk assessment. Using analytical simulations with HotSpot Health Physics software, estimates were made for individuals continuously within the 700 mSv isodose zone, a dose also linked to immediate deterministic effects such as acute radiation syndrome. To improve operational efficiency and resource use in Chemical, Biological, Radiological, and Nuclear scenarios, the study discusses recommended engineering countermeasures, including optimizing route planning and protective barriers. It introduces a new proposed risk factor model that aims to measure the effects of overlapping radioactive plumes caused by changing local wind directions, highlighting increased vulnerability in transition zones. Unlike previous military or civilian radiological risk studies, this research focuses on the implications of plume interactions in real-time tactical environments, offering a useful framework for enhancing the safety and effectiveness of combat engineering operations facing radiological threats in hostile settings.
- 4 August 2025
- Published by Taylor & Francis in International Journal of Radiation Biology
- Vol. 101 (9) , 982-990
Evaluating the health impacts of two extreme-releasing radiological hypothetical events
This simulation study compares the long-term health for military troops exposed by a drone dispersing low-dose radioactive material either in a ground-level explosion or aerial release. It examines the consequences of dispersing radioactive material using a drone through both ground-level explosions and aerial releases. The simulations were conducted over four days, without adjusting for radioactive decay, and considered six atmospheric stability classes. The findings indicate that ground-level explosions lead to higher total effective dose equivalent (TEDE) values across all atmospheric stability classes, especially in stable atmospheric conditions (Classes E and F). The excess relative risk of leukemia induced by radiation is also more significant in the ground-level explosion scenario, particularly for younger individuals (20 years old). The study's findings indicate that ground-level dispersion of substances poses a greater risk to public health compared to drone-based dispersal methods. This risk is notably exacerbated under stable atmospheric conditions and is particularly pronounced among younger demographics. Such outcomes hold significant implications for policy formulation and strategic decision-making within the context of public health initiatives and military radiological operations.
- 4 August 2025
- Published by Taylor & Francis in International Journal of Radiation Biology
- Vol. 101 (9) , 982-990
Evaluating the health impacts of two extreme-releasing radiological hypothetical events
This simulation study compares the long-term health for military troops exposed by a drone dispersing low-dose radioactive material either in a ground-level explosion or aerial release. It examines the consequences of dispersing radioactive material using a drone through both ground-level explosions and aerial releases. The simulations were conducted over four days, without adjusting for radioactive decay, and considered six atmospheric stability classes. The findings indicate that ground-level explosions lead to higher total effective dose equivalent (TEDE) values across all atmospheric stability classes, especially in stable atmospheric conditions (Classes E and F). The excess relative risk of leukemia induced by radiation is also more significant in the ground-level explosion scenario, particularly for younger individuals (20 years old). The study's findings indicate that ground-level dispersion of substances poses a greater risk to public health compared to drone-based dispersal methods. This risk is notably exacerbated under stable atmospheric conditions and is particularly pronounced among younger demographics. Such outcomes hold significant implications for policy formulation and strategic decision-making within the context of public health initiatives and military radiological operations.
- 18 July 2025
- Published by Taylor & Francis in Nuclear Science and Engineering
- Vol. 200 (5) , 1263-1271
Estimating the Impact of an Electromagnetic Pulse Caused by a Hypothetical Surface-Level Nuclear Explosion on Urban Critical Infrastructure
- 18 July 2025
- Published by Taylor & Francis in Nuclear Science and Engineering
- Vol. 200 (5) , 1263-1271
Estimating the Impact of an Electromagnetic Pulse Caused by a Hypothetical Surface-Level Nuclear Explosion on Urban Critical Infrastructure
- Open Access
- 4 July 2025
- Published by Sociedade Brasileira de Proteção Radiológica - SBPR in Brazilian Journal of Radiation Sciences
- Vol. 13 (3) , e2862
Intentional radiological exposure scenario evaluation by comparisons based on computer simulation
Nuclear or radiological mass incidents represent a threat that requires sophisticated coping strategies. This study aims to contribute by presenting a dual computational modeling methodology that juxtaposes numerical and analytical models to address a specific radioactive release scenario. This methodology seeks to extend beyond the theory underlying the modeling processes, favoring decision-making, especially in the early stages of such confrontation. This investigation applied a dual-model structure based on numerical methods and analytical techniques to simulate a radiological scenario promoted by activating a radiological dispersal device (RDD). It is important to emphasize that this methodology is not limited to RDD scenarios and is being proposed for application to any external release of radioactive materials. By evaluating and comparing the results of the simulations, particularly in areas close to the release point and in shorter time intervals, it is possible to verify the most appropriate model and identify scenarios in which the two models produce convergent results. The findings highlight the importance of estimating radiation doses, suggesting that such estimates can influence the understanding of radiological risks and their dependence on local atmospheric conditions. Careful interpretation and application of such results can mitigate epidemiological risks, enhance coordination capabilities, and stimulate the development of strategic responses.
- 4 July 2025
- Published by Sociedade Brasileira de Proteção Radiológica - SBPR in Brazilian Journal of Radiation Sciences
- Vol. 13 (3) , e2862
Open AccessIntentional radiological exposure scenario evaluation by comparisons based on computer simulation
Nuclear or radiological mass incidents represent a threat that requires sophisticated coping strategies. This study aims to contribute by presenting a dual computational modeling methodology that juxtaposes numerical and analytical models to address a specific radioactive release scenario. This methodology seeks to extend beyond the theory underlying the modeling processes, favoring decision-making, especially in the early stages of such confrontation. This investigation applied a dual-model structure based on numerical methods and analytical techniques to simulate a radiological scenario promoted by activating a radiological dispersal device (RDD). It is important to emphasize that this methodology is not limited to RDD scenarios and is being proposed for application to any external release of radioactive materials. By evaluating and comparing the results of the simulations, particularly in areas close to the release point and in shorter time intervals, it is possible to verify the most appropriate model and identify scenarios in which the two models produce convergent results. The findings highlight the importance of estimating radiation doses, suggesting that such estimates can influence the understanding of radiological risks and their dependence on local atmospheric conditions. Careful interpretation and application of such results can mitigate epidemiological risks, enhance coordination capabilities, and stimulate the development of strategic responses.