Development of emergency response systems by intelligent and integrated approaches for marine oil spill accidents
Bibliographic record
Abstract
Oil products play a pervasive role in modern society as one of the dominant energy fuel sources. Marine activities related to oil extraction and transportation play a vital role in resource supply. However, marine oil spills occur due to such human activities or harsh environmental factors. The emergency accidents of spills cause negative impacts on the marine environment, human health, and economic loss. The responses to marine oil spills, especially large-scale spills, are relatively challenging and inefficient due to changing environmental conditions, limited response resources, various unknown or uncertain factors and complex resource allocation processes. The development of previous research mainly focused on single process simulation, prediction, or optimization (e.g., oil trajectory, weathering, or cleanup optimization). There is still a lack of research on comprehensive and integrated emergency responses considering multiple types of simulations, types of resource allocations, stages of accident occurrence to response, and criteria for system optimizations. Optimization algorithms are an important part of system optimization and decision-making. Their performance directly affacts the quality of emergency response systems and operations. Thus, how to improve efficiency of emergency response systems becomes urgent and essential for marine oil spill management. The power and potential of integrating intelligent-based modeling of dynamic processes and system optimization have been recognized to better support oil spill responders with more efficient response decisions and planning tools. Meanwhile, response decision-making combined with human factor analysis can help quantitatively evaluate the impacts of multiple causal factors on the overall processes and operational performance after an accident. To address the challenges and gaps, this dissertation research focused on the development and improvement of new emergency response systems and their applications for marine oil spill response in the following aspects: 1) Realization of coupling dynamic simulation and system optimization for marine oil spill responses - The developed Simulation-Based Multi-Agent Particle Swarm Optimization (SA-PSO) modeling investigated the capacity of agent-based modeling on dynamic simulation of spill fate and response, particle swarm optimization on response allocation with minimal time and multi-agent system on information sharing. 2) Investigation of multi-type resource allocation under a complex simulation condition and improvement of optimization performance - The improved emergency response system was achieved by dynamic resource transportation, oil weathering and response simulations and resource allocation optimization. The enhanced particle swarm optimization (ME-PSO) algorithm performed outstanding convergence performance and low computation cost characteristics integrating multi-agent theory (MA) and evolutionary population dynamics (EPD). 3) Analysis and evaluation of influencing factors of multiple stages of spill accidents based on human factors/errors and multi-criteria decision making - The developed human factors analysis and classification system for marine oil spill accidents (HFACS-OS) framework qualitatively evaluated the influence of various factors and errors associated with the multiple operational stages considered for oil spill preparedness and response (e.g., oil spill occurrence, spill monitoring, decision making/contingency planning, and spill response). The framework was further coupled with quantitative data analysis by Fuzzy-based Technique for Order Preference by Similarity to Idea Solution (Fuzzy-TOPSIS) to enhance decision-making during response operations under multiple criteria. 4) Development of a multi-criteria emergency response system with the enhanced optimization algorithm, multi-mode resource transportation and allocation and a more complex and realistic simulation modelling - The developed multi-criteria emergency response system (MC-ERS) system integrated dynamic process simulations and weighted multi-criteria system optimization. Total response time, response cost and environmental impacts were regarded as multiple optimization goals. An improved weighted sum optimization function was developed to unify the scaling and proportion of different goals. A comparative PSO was also developed with various algorithm-improving methods and the best-performing inertia weight function. The proposed emergency response approaches in studies were examined by oil spill case studies related to the North Atlantic Ocean and Canada circumstances to analyze the modelling performance and evaluate their practicality and applicability. The developed optimization algorithms were tested by benchmarked functions, other optimization algorithms, and an oil spill case. The developed emergency response systems and the contained simulation and optimization algorithms showed the strong capability for decision-making and emergency responses by recommending optimal resource management or evaluations of essential factors. This research was expected to provide time-efficient, and cost-saving emergency response management approaches for handling and managing marine oil spills. The research also improved our knowledge of the significance of human factors/errors to oil spill accidents and response operations and provided improved support tools for decision making. The dissertation research helped fill some important gaps in emergency response research and management practice, especially in marine oil spill response, through an innovative integration of dynamic simulation, resource optimization, human factor analysis, and artificial intelligence methods. The research outcomes can also provide methodological support and valuable references for other fields that require timely and effective decisions, system optimizations, process controls, planning and designs under complicated conditions, uncertainties, and interactions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".