ABSTRACT Practices developed under war are increasingly offered as lessons for critical infrastructure operators in peacetime, yet they are rarely examined for what made them work. This article analyzes 14 decisions taken between 2022 and 2024 by the municipal enterprise responsible for Kyiv's city network, cybersecurity and digital services, and asks which of them transfer to municipal and utility operators in the United States. Drawing on the policy transfer literature, it develops a framework that separates each practice from the conditions that enabled it: emergency authority, public tolerance of disruption, procurement speed, and structural assets that a given city may or may not have. Each decision is classified as transferable as is, transferable with modification, or not transferable. Three of the 14 transfer as is, nine transfer only with modification, and two do not transfer. The main finding is that the practices themselves are rarely the obstacle. What does not travel are the enablers, which war supplied on demand and which a city in peacetime must arrange in advance: pre‐delegated decision authority, standing contracts and pre‐identified sites, and data‐sharing agreements with utilities. Because recent federal policy places more responsibility for preparedness on state and local governments, these arrangements are largely within the reach of city governments themselves.
ABSTRACT The field of network security has greatly advanced due to the implementation of intrusion detection systems (IDS), which are capable of identifying unforeseen threats within network traffic. Contemporary techniques for detecting anomalies in networks predominantly utilize traditional machine learning models. The proposed Enhanced Fault‐Tolerant Transformer‐Based Slimmable Pruned Neural Network with Snow Geese Optimization (EFT‐SPNet‐SGO) to filter network intrusion is integrated with OPBFTCM based authentication, statistical decimal scaling normalization (SDSN), scale‐aware modulation meet transformer (SMMT) with feature extraction and slimmable pruned neural network (SPNet) with classification using snow geese optimization (SGO). The OPBFTCM is adopted for the authentication process. The data is then preprocessed using SDSN. Then, relevant features are extracted using the SMMT. Finally, the detection and categorization of network intrusions is carried out by the SPNet, which is optimized using the SGO algorithm. The efficacy of the proposed strategy can be assessed using metrics such as recall, accuracy, precision, F1‐score, latency, and comparability to other approaches. The experiment's findings demonstrate that the designed model outperformed, with precision, f‐measure, accuracy, and recall values obtained from the network security laboratory‐knowledge discovery and data mining (dataset) all reaching 99.3%.
ABSTRACT Oil spills endanger the environment and require immediate detection and proper technology for evaluation and containment. These have caused significant drawbacks to conventional detection methods in terms of resources, time, and especially limitations in scale over the extent of oceanic areas. Most conventional procedures include disturbance and aerial surveillance methods with very long‐time frames for getting evidence and broad resource allocations. Such anomalies respond very late and cover relatively small areas, making them very ineffective in spill monitoring. In this study, we present a new adaptive oil spill detection framework, Wavelet‐ Long Short ‐Term Memory (LSTM) with Attention, for smart ocean systems using the Internet of Things (IOT) and real‐time satellite data. The model involves a wavelet transform to reduce noise and extract features, an LSTM so as to model the spatiotemporal data of the environment, and an attention mechanism that will help tend toward the critical features. The whole system is optimized using the Adam algorithm, which ensures efficient learning and quicker convergence. On the basis of the results of the experiment, the model turns out to be highly efficient in terms of attaining accuracy with the following figures: 98.95%, precision: 98.45%, recall: 98.52%, and F1 score: 98.00%. Additionally, the present study confirms the efficiency of the model in terms of several other assessment criteria, namely, ROC and Precision‐Recall curves, which confirm good generalization and low rates of misclassification. Therefore, the proposed system can be employed for the purpose of autonomous, scalable, and efficient monitoring of environment protection and disaster management in marine ecosystems. The suggested approach is known as the Wavelet Transform, LSTM and Attention Mechanism.
ABSTRACT The integration of Internet of Things (IoT) with cloud platforms has generated an extremely decentralized, data‐driven network that is transforming industries. At the same time, the integration poses intricate cybersecurity challenges because of scalability requirements, device heterogeneity, and an evolving threat environment. Conventional security metrics are inadequate in some circumstances, creating a necessity for adaptive, intelligent, and proactive defense modules. Artificial intelligence (AI) offers beneficial capabilities for intrusion detection, predictive threat modeling, anomaly detection, and malware analysis. Moreover, deep learning (DL) models have proved exceptional effectiveness in automatically learning intricate patterns from larger‐scale network traffic information, allowing precise identification of advanced cyberattacks in IoT‐cloud systems. Therefore, this manuscript presents a Fine‐Grained Hybrid Attention‐Aware Deep Learning Framework for Multi‐Layer Cyberattack Detection (FGHADL‐MLCD). The major intention of the proposed model is to enhance the reliability and intelligence of cyberattack detection in large‐scale IoT‐cloud infrastructures. The FGHADL‐MLCD approach operates through a sequence of stages, including preprocessing, attribute selection, cyberattack classification, and model optimization. Initially, the collected network traffic data from TON‐IoT and Edge‐IIoT datasets undergo preprocessing operations such as data cleaning, label encoding, feature scaling, and dataset splitting to ensure consistency of input data. Following that, an effective attribute selection mechanism is introduced using the Boruta feature selection technique, which identifies the most relevant network traffic features. For multi‐layer cyberattack detection and classification, the proposed model employs a bidirectional temporal convolutional network integrated with an attention mechanism. Furthermore, the training process of the deep learning model is optimized using the Adam optimizer. To assess the model's effectiveness, extensive experiments are conducted using the benchmark IoT cybersecurity datasets. The experimental results indicate that the proposed technique achieves improved cyberattack detection performance compared to recent models. Thus, the proposed model delivers a reliable solution for securing IoT‐cloud infrastructures against diverse and evolving cyber threats.
ABSTRACT The strategic siting of shore bases, onshore facilities that support offshore energy operations, is traditionally approached as a problem of logistical efficiency, emphasizing distance, port capacity, and transport infrastructure. However, in emerging offshore frontiers, decision‐makers must simultaneously balance logistical, social, and local content considerations that are difficult to incorporate into conventional location analysis. The objective of this paper is to quantify how these dimensions are traded off against one another by operationalizing shore base location decisions within a discrete choice framework. A stated preference experiment was conducted with 60 professionals from Colombia's offshore oil and gas sector, each completing eight binary choice tasks (480 observed choices), and analyzed through Multinomial Logit and Hierarchical Bayes models. Port infrastructure emerged as the most influential attribute: its absence carries a trade‐off equivalent to 385 km of additional distance. Yet social conflict outweighed several traditional logistical considerations once explicit trade‐offs were introduced, with respondents willing to travel on average approximately 250 km further to operate in areas of low social conflict, whereas local content attributes exerted limited influence on siting decisions. A preference‐based segmentation via hierarchical clustering identified four distinct strategic profiles, revealing substantial heterogeneity behind these averages. By demonstrating how discrete choice methods can operationalize infrastructure location decisions that jointly involve engineering, social, and local content criteria, this study advances the facility location literature and provides a transferable framework for conflict‐sensitive planning in emerging offshore energy regions.
