Multi-Agent Safety Monitor

Condition monitoring or safety monitor is a cost-effective safety protection technique which is widely used in safety-critical industries, such as nuclear plants and aerospace operation. The adoption of safety monitor offers continuous or real-time safety protection at a high level of diversity and its implementation is relatively simple and thus cost-effective for safety verification and validation. In practice, the safety monitors are practical to improve the overall system safety because: 1) Safety monitors are generally small and do not normally have complex algorithms; 2) Safety monitors normally involve checking the essential safety features/properties of the large safety-critical railway systems during their operations; 3) Safety monitors are often independent of existing systems. It can be taken as an add-on safety feature and it is physically and functionally separated from the existing systems, which boosts the reliability of the system. In general, the safety monitors of safety-critical railway systems must achieve the same level or higher level of the integrity and safety required by the monitored systems. Formal method approach provides a traceable and accountable approach to understand, plan and manage safety of safety-critical railway systems, and is an effective technique to improve the overall system safety and reliability. Safety monitors normally have many safety properties in logic condition forms need to be satisfied, and it is the domain that formal methods are good at and should be applied to. This paper discusses a development process to utilise the formal methods to improve the safety and quality of safety critical railway systems. A simple example on railway signalling is then given to demonstrate the application of formal methods in safety-critical transport system.

The objectives of this research are to find the proper measures to improve efficiency of Thai-Lao cross border transportation (R9 route) for Thai shippers and study their performance after implementing the developed measures. This research implements both quantitative and qualitative research, where the population of this research consists of shipper’s groups which transport products via the Thai Lao border located along R9 route. The research tools are questionnaire and structured interview form passed variability and reliability tests. Then the questionnaires are distributed to 3 large size firms (sample size of 316 people) and an in-depth interview with 18 people. This research analyzes gathered data using descriptive analysis, construct validity, structural equation modeling (SEM), path analysis, and content analysis for qualitative data. The analysis results reveal important findings according to each research objective as follows. For the first objective, to improve cross-border transportation, the case study firms should concentrate on three measures as: 1) transportation safety in terms of sound safety control and monitor, increasing rest areas and truck stops including those on pavement or roadside, safety monitoring during transportation, and weather conditions checking; 2) documentation procedure in terms of reducing the procedure steps, IT implementation, and procedure improvement both inbound and outbound; and 3) vehicles management in terms of appropriate resources allocation, facilities availability, and appropriate resources selection. For the second objective, the study of the case study firms’ performance after implementing the developed measures, it reveals benefits in several aspects of 1) cheaper (they can deliver items with lower costs, lower expenses, or lower price than before implementation as well as other competitors); 2) faster (better customers responsiveness, faster delivery and operating than other competitors, and lead time reduction), and 3) better (in operations with less bottlenecks, mistakes, and disruptions manifested in delivery; higher competitiveness and better service quality). To summarize, the implementation of the developed model can help the shippers to increase their capability as well as their competitiveness.

Machine learning (ML) provides no guarantee of safe operation in safety-critical systems such as autonomous vehicles. ML decisions are based on data that tends to represent a partial and imprecise knowledge of the environment. Such probabilistic models can output wrong decisions even with 99% of confidence, potentially leading to catastrophic consequences. Moreover, modern ML algorithms such as deep neural networks (DNN) have a high level of uncertainty in their decisions, and their outcomes are not easily explainable. Therefore, a fault tolerance mechanism, such as a safety monitor (SM), should be applied to guarantee the property correctness of these systems. However, applying an SM for ML components can be complex in terms of detection and reaction. Thus, aiming at dealing with this challenging task, this work presents a benchmark architecture for testing ML components with SM, and the current work for dealing with specific ML threats. We also highlight the main issues regarding monitoring ML in safety-critical environments.

Workplace safety is a critical concern in industrial environments, where the risk of accidents and injuries is inherently high. Traditional safety monitoring methods often rely on manual inspections and periodic checks, which can be inefficient and prone to human error. This project introduces the Industrial Safety Monitor, a comprehensive application designed to enhance workplace safety through advanced monitoring techniques using state-of-the-art object detection and classification models

Based on the north anchorage foundation of Runyang Bridge, some beneficial results were obtained concerning the stability of Anchorage foundation. The finite element method shows great advantage to analysis the nonlinear problem concerning the anchorage foundation stability, which can raise the efficiency and calculation accuracy greatly. The safety monitor model, based on the output of three-dimension finite element simulation, is reasonable. The dynamic prediction analyses based on the monitoring data ensured the stability and safety of the anchorage foundation.

This paper discusses integrated safety monitoring system design of in-flight aircraft. During a flight mission, the aircraft, the operation of crew members and the dynamic change of flight environments interacts as a system. The change of safety situation of an aircraft should be attributed to the behavior of those human-machine-environment factors, which established the basis of the in-flight safety monitor system designing.

Integration of diagnostics with a memory built-in self-test (BIST) allowing both an effective and efficient manufacturing test as well as an effective diagnostic capability is detailed. Detection of failures within a memory are the primary objective of an embedded memory BIST. Inclusion of comprehensive diagnostics, to isolate faults incurred during the manufacturing or design process, are an extension to the memory BISI, further utilizing the existing circuitry and functionality. Detection and diagnosis of failures within an embedded memory can be realized throughout the entire manufacturing process.

Architecture Analysis and Design Language (AADL) is an emerging industry standard notation for modeling architecture of embedded real-time systems. AADL has capability to model normal and faulty behavior of the system. It can model error behavior, for all of its components, their interactions and propagations. This capability makes it suitable for modeling safety critical systems, where fault detection, isolation and re-configuration are of paramount importance. This paper proposes using AADL to do safety validation of an embedded real-time system. AADL is used as a basis to systematically derive requirements and parameters specific to safety validation at both platform level and application level. Behavior and error models corresponding to system level AADL models are used to derive the parameters. Such parameters can then be monitored at the system integration testing level automatically using a safety monitor. We present our ideas using a detailed case study involving safety validation of a automatic flight control system (AFCS) from its AADL model that includes behavior and error modeling.

This paper presents a technique for failure detection and diagnosis (FDD) for on-board vehicle applications. Failures are represented by abrupt changes in the characteristic parameters of a system that occur at random instants of time. This representation facilitates the formulation of the FDD problem as a combined state and parameter estimation problem of a system with uncertain parameters. This non-linear estimation problem is solved using a Bayesian approach and implementing a suboptimal filter. The performance of the FDD system is demonstrated in simulations of an active suspension subjected to sensor failures. In the simulations, the system diagnosed failures correctly within .05 seconds from their occurrence.

Recent advances in microelectronic circuit have enabled the application of process diagnostics to a variety of systems to improve performances and the reliability. Failure detection and isolation strategies monitor a system for degradations and if detected, classify the failure source. One of the most important methods for failure detection and diagnosis is the analysis of the variations of the estimated parameter of DC Motor. In this paper is presented a structure of a method for failure detection and diagnosis based on continuous time parameter estimation of a mathematical model of the DC Motor. Continuous time parameters can be related easily to the physical characterizations of DC Motor. The parameter is directly estimated by well-known prediction error method of identification for dynamic systems. The failure detection it is based on changing the process coefficient and statistical decision methods. Unlike usual papers, the failure can be detected whether others process parameters are not constant.

Steady-state visual evoked potential (SSVEP)-based brain–computer interface (BCI) technology offers a promising solution for wheelchair control by translating neural signals into navigation commands. A major challenge—signal noise caused by eye blinks—is addressed in this feasibility study through real-time blink detection and correction. The proposed design utilizes sensors to capture both SSVEP and blink signals, enabling the isolation and compensation of interference, which improves control accuracy by 14.68%. Real-time correction during blinks significantly enhances system reliability and responsiveness. Furthermore, user data and global positioning system (GPS) trajectories are uploaded to the cloud via Wi-Fi 6E for continuous safety monitoring. This approach not only restores mobility for users with physical disabilities but also promotes independence and spatial autonomy.

Photovoltaic (PV) technology allows large-scale investments in a renewable power-generating system at a competitive levelized cost of electricity (LCOE) and with a low environmental impact. Large-scale PV installations operate in a highly competitive market environment where even small performance losses have a high impact on profit margins. Therefore, operation at maximum performance is the key for long-term profitability. This can be achieved by advanced performance monitoring and instant or gradual failure detection methodologies. We present in this paper a combined approach on model-based fault detection by means of physical and statistical models and failure diagnosis based on physics of failure. Both approaches contribute to optimized PV plant operation and maintenance based on typically available supervisory control and data acquisition (SCADA) data. The failure detection and diagnosis capabilities were demonstrated in a case study based on six years of SCADA data from a PV plant in Slovenia. In this case study, underperforming values of the inverters of the PV plant were reliably detected and possible root causes were identified. Our work has led us to conclude that the combined approach can contribute to an efficient and long-term operation of photovoltaic power plants with a maximum energy yield and can be applied to the monitoring of photovoltaic plants.

Submodule (SM) open-circuit failures severely affect the reliable operation of modular multilevel converter (MMCs). A comprehensive SM open-circuit failure diagnosis process includes three steps with failure detection, localization, and classification. However, the existing diagnosis methods perform these steps separately, which makes the diagnosis process complicated. Furthermore, the efficiency is relatively low with separate diagnosis steps. To address this issue, a diagnosis method with variance measurement is proposed in this paper to combine failure detection, localization, and classification. The proposed method makes use of two fault characteristics of SM open-circuit failures: 1) SM capacitor voltages have distinct distributions upon different faulty IGBTs; 2) Numerous SMs in the MMC make the failure diagnosis equivalent to statistical outlier analysis. On account of the fault characteristics, a variance-based index called VAR is proposed to evaluate the distributions of SM capacitor voltages. The VAR values are then analyzed with quartile analysis, which is a statistical outlier analysis method, to complete the diagnosis process. The proposed failure diagnosis method features high reliability, high compatibility, and high cost-effectiveness. Finally, the feasibility of the proposed method is verified through both simulations and experiments under various failure scenarios.

As the safety analyses of critical systems typically cease or reduce in their utility after system certification useful knowledge about the behaviour of the system in conditions of failure remains unused in the operational phase of the system lifecycle. In this paper, we show that this knowledge could be usefully exploited in the context of an on- line hazard-directed monitoring scheme in which a suitable specification derived from design models and safety analyses forms a reference monitoring model. As a practical application of this approach, we propose a safety monitor that can operate on such models to support the on-line detection, diagnosis and control of hazardous failures in real-time. We discuss the development of the monitoring model and report on a case study that we performed on a laboratory model of an aircraft fuel system.

BackgroundN-terminal pro-brain natriuretic peptide (NT-proBNP) is a valuable biomarker for diagnosis and prognosis of heart failure in adults, included into the European Society Guidelines for heart failure (2016).1 It is...