Design of Models and Algorithms for Decision Support in Aviation Safety Tasks

Evaluation of decision making support models for managing re-equipping fire truck park using statistical testing method

Purpose. The article examines the possibility of applying the statistical testing method to evaluate the effectiveness of decision making support models for distributing fire trucks among territorial fire and rescue garrisons. Based on previously developed criteria of operational readiness, technical readiness, and modern equipment availability, a probabilistic approach to analyzing the stability of management decisions is proposed. A numerical example of ranking garrisons with assessment of confidence intervals and probabilities of making correct decisions is provided. Methods. The methodology is based on the statistical testing method, which accounts for the probabilistic nature of initial data. The study includes a mathematical apparatus of simulation modeling — generation of random variables with specified distribution laws (normal, exponential, Erlang), procedures for multi-criteria optimization using the Cobb–Douglas function, and subsequent statistical processing of results to construct confidence intervals and estimate probabilities. Findings. As a result of the modeling, probabilistic estimates of ranking five garrisons using two-criteria and three-criteria models were obtained. It was established that the three-criteria model, which accounts for modern equipment availability, provides greater stability to variations in input data and a lower probability of error. Confidence intervals for criterion values and probabilities of objects falling into priority groups (“red,” “yellow,” "green") were determined. The Kendall concordance coefficient was calculated, amounting to W2 = 0.83 for the two-criteria model and W3 = 0.91 for the three-criteria model, indicating high ranking consistency. Research application field. The proposed methodology is intended for use in the logistics system of EMERCOM of Russia. The results can be used by decision-makers to improve justifiability and reduce risks when distributing limited resources (fire trucks) among territorial fire and rescue garrisons under conditions of data uncertainty. Conclusions. The application of the statistical testing method has proven effective for comparative analysis of decision support models under uncertainty. The three-criteria model is recognized as more preferable for practical use, as it reduces the risk of erroneous classification of garrisons and provides more stable ranking.

Uncertainty-Aware Automated Maintenance Decision-Making for Elevator Mechatronic Systems

To address the fundamental challenge of translating probabilistic health assessments into reliable maintenance decisions for elevator mechatronic systems, this paper proposes the first uncertainty-aware dual-threshold optimization framework that systematically bridges Bayesian perception with optimal decision-making. Leveraging posterior health distributions (mean and variance) output by the Bayesian Fusion for Elevator Health Assessment, the method constructs supervised learning pairs from multi-year historical records. A three-objective composite loss function-integrating matching accuracy, uncertainty coverage, and decision separability-is optimized via Bayesian optimization to calibrate optimal thresholds. Critically, the 95% confidence interval is explicitly embedded as a constraint, ensuring high-uncertainty equipment remains safely distant from decision boundaries. Validated on one-year operational data from 100 elevators, the framework achieves 84.0% matching accuracy (17.0% improvement), 0.71 maintenance cost index, and 0.12 safety risk index, delivering Pareto-optimal trade-offs among decision accuracy, maintenance economy, and operational safety. The offline learning and online deployment paradigm requires no additional hardware and integrates directly into existing supervision platforms.

Modeling and Optimization of Complex Civil Aviation System

The civil aviation system is a dynamic and complex system involving airlines, flights, airlines, airports and runways. There are many factors affecting the safety and efficiency of civil aviation system. Under the circumstance of limited resources, how to design, optimize and evaluate the resource allocation of civil aviation system will greatly improve its management science level. However, if the scheduling scheme of civil aviation system is only prepared manually, it is not only time-consuming and laborious, but also the experience scheme is hardly optimal. With the help of big data and artificial intelligence, it is possible to generate the design and optimization scheme of civil aviation system by using computers. The main goal of this book is to build mathematical models and design algorithms for several common types of civil aviation problems from an operations research perspective. Finally, an example is given to verify the validity of the model and algorithm. Another goal of this book is how to evaluate the operational efficiency of the civil aviation system, where the different objects would be scored based on the establishment of evaluation indicators. Finally, an example is given to verify the validity of this study.

A Comparative Study of Decision-Support Approaches for Managing Aircraft Landing Gear Systems Reliability

Aircraft landing gear systems are essential to safe and efficient flight operations, as they operate under demanding conditions during takeoff, landing, and ground maneuvers. Because they are exposed to high loads and repeated stress, their reliability has a direct impact on safety, maintenance planning, and overall operational performance. This study explores the use of three decision-support approaches, namely the Analytic Network Process (ANP), the Decision-Making Trial and Evaluation Laboratory (DEMATEL), and an adapted Fuzzy Cognitive Map (FCM), to examine a critical landing gear subsystem. The analysis identifies the most influential engineering management factors affecting subsystem reliability and compares how the three methods capture factor importance and interdependencies. The results show that maintenance planning efficiency, failure mode criticality, cost of downtime, and workforce coordination consistently emerge as key reliability-related dimensions. The adapted FCM provides a broader uncertainty-aware representation of indirect causal propagation, while ANP supports prioritization through network-based weighting and DEMATEL highlights cause–effect structures among the factors. Overall, the findings demonstrate that combining these complementary methods can strengthen reliability-oriented decision-making by clarifying which factors should receive priority in maintenance, operational planning, and management strategies.

ADAPTIVE LEARNING OF FAULT‑TOLERANT SOFTWARE ENGINEERING FOR UAV AND NANOSATELLITE ONBOARD COMPUTING

Coverage-Driven Verification for Safety-by-Design in AI-Based Collision Avoidance Systems

This work presents a method for representativeness assessment of AI/ML constituent ODDs in the context of aviation safety assurance and illustrates how statistical distribution comparison methods can support the assessment of representativeness for safety-critical AI applications.