Dynamic human reliability and dual-type quality prediction in human-machine collaborative tuning manufacturing systems

Los sistemas de recomendación han tenido un rápido crecimiento tanto en investigación académica como en aplicaciones industriales en los últimos años. Las recomendaciones se nos proporcionan cuando elegimos elementos como música, películas, restaurantes, noticias, etc. Como parte de las técnicas de recomendación más prometedoras, los sistemas basados en recomendación de filtros colaborativos desempeñan un papel importante en las aplicaciones actuales. La aparición de nuevas fuentes de información requiere la combinación de sistemas de recomendación con técnicas avanzadas de otros campos, como el análisis de redes sociales, el procesamiento del lenguaje natural, el aprendizaje en profundidad, etc. Esta combinación mejora la calidad de las recomendaciones. Los sistemas de recomendación anteriores se centran principalmente en las mejoras de la precisión de la recomendación. Recientemente, surge una tendencia para la evaluación del rendimiento de los sistemas de recomendación que toma muchas propiedades nuevas en consideración. La fiabilidad es uno de los aspectos más notables y será profundamente estudiado en esta tesis. Se han propuesto algunas medidas para medir la fiabilidad de los sistemas de recomendación de filtrado colaborativo basados en la memoria. Sin embargo, observamos un área sin cubrir en el diseño de las medidas de fiabilidad adecuadas para los sistemas de recomendación de filtros colaborativos basados en modelos, especialmente aquellos basados en la factorización matricial. También encontramos que hay una falta de medidas de calidad de fiabilidad adecuadas. Esta tesis se concentra en proporcionar soluciones a ambas ausencias. Después de un estudio exhaustivo de trabajos de investigación del estado del arte, proponemos una nueva medida de fiabilidad que supera los desafíos de generalización y abstracción causados por los factores latentes de los métodos de factorización matricial. La medida propuesta se compara con con métodos y algoritmos representativos, usando bases de datos publicas, tanto académicas como comerciales. Los resultados experimentales muestran que nuestra nueva medida logra mejoras en la calidad de la recomendación. Además, la medida de fiabilidad propuesta también tiene un impacto muy positivo contra los "Shilling Attacks". Para obtener una mejora respecto a las medidas existentes de calidad de la fiabilidad, también proponemos, hasta donde sabemos, la primera medida de calidad de fiabilidad universal que es capaz de proporcionar valores numéricos para cualquier tipo de sistema de recomendación. Esta medida también se ha probado usando conjuntos de datos públicos reales y ha sido comparada con los métodos de calidad de la fiabilidad existentes. Basándonos en los resultados de los experimentos, concluimos que nuestra medida supera los últimos trabajos de investigación, proporcionando valores de calidad de fiabilidad más precisos e informativos. ----------ABSTRACT---------- Recommender systems have experienced a rapid growth in both academical research and industrial applications in latest years. Recommendations are provided to us when we choose items such as music, movies, restaurants, news, etc. As the most promising recommendation techniques, collaborative filtering recommender systems play an important role in current applications. The emergence of novel information sources requires to combine recommender systems with several advanced techniques coming from other fields like social network analysis, natural language processing, deep learning etc. This combination improves the quality of recommendations. Previous recommender systems research mainly focuses on the improvements of recommendation accuracy. Recently a new trend arises for the performance evaluation of recommender systems that takes many novel properties into consideration. Reliability is one of the most remarkable aspects, which will be deeply studied in this thesis. Some measures have been proposed for measuring reliability of memory based collaborative filtering recommender systems. However, we observe a gap in the design of proper reliability measures for model based collaborative filtering recommender systems, especially those based on matrix factorization. We also found that there is a lack in proper reliability quality measures. This thesis concentrates on providing solutions to both lacks. After a comprehensive study of state-of-the-art research works, we propose a novel reliability measure that overcomes the challenges of generality and abstraction caused by latent factors of matrix factorization methods. The proposed measure is tested against current baselines on public academic and commercial datasets; experiment results show that our novel measure achieves better improvements on recommendation quality. What’s more, the proposed reliability measure also has a very positive impact against shilling attacks. In order to fill in the blank of reliability quality measures, we also propose, as far as we know, the first universal reliability quality measure that is able to provide numerical values for any type of recommender systems. This measure has also been tested on public academic and commercial datasets using current methods. Based on the experiment results we conclude that our quality measure outperforms state-of-the-art research works by providing more accurate and informative reliability quality values.

A bibliometric analysis of human reliability research

Offshore system safety and reliability considering microbial influenced multiple failure modes and their interdependencies

As a popular means for capturing behavioural requirements, scenarios show how components interact to provide system-level functionality. If component reliability information is available, scenarios can be used to perform early system reliability assessment. In previous work we presented an automated approach for predicting software system reliability that extends a scenario specification to model (1) the probability of component failure, and (2) scenario transition probabilities. Probabilistic behaviour models of the system are then synthesized from the extended scenario specification. From the system behaviour model, reliability prediction can be computed. This paper complements our previous work and presents a sensitivity analysis that supports reasoning about how component reliability and usage profiles impact on the overall system reliability. For this purpose, we present how the system reliability varies as a function of the components reliabilities and the scenario transition probabilities. Taking into account the concurrent nature of component-based software systems, we also analyse the effect of implied scenarios prevention into the sensitivity analysis of our reliability prediction technique.

A novel dynamic function allocation method in human-machine systems focusing on trigger mechanism and allocation strategy

The dramatic increase in the number of heterogeneous computing architectures (HCAs) which include CPUs, GPUs, TPUs, FPGAs, and new neuromorphic processors has transformed the high-performance computing (HPC) and artificial intelligence (AI) workloads. As the complexity of architecture increases there is corresponding uncertainty on reliability and performance checks, particularly when systems are incorporated with different processing elements, interconnections, and memory levels. Modern non-deterministic or non-stochastic heterogeneous environments are associated with traditional reliability models that feel great necessity to model nonlinear interdependencies between components. The current paper proposes an innovative AI-based modeling framework that would be capable of measuring system reliability and performance indicators in the context of HCAs. The suggested solution merges Deep Neural Networks (DNNs), Bayesian Inference Models, and the strategies of Reinforcement Learning (RL) to forecast the increase of the system reliability, the performance bottlenecks, and the mean time between failures (MTBF) at the system level. To train AI models that can capture the multi-domain interactions of computation, communication, and thermal dynamics, a large-scale simulation dataset was created with synthetic benchmarks, actual workloads (SPEC ACCEL, MLPerf, HPCG), unlikely fault profiles, and injected fault profiles to have AI models that are trained to simulate interactions in large-scale systems. The approach puts the feature engineering of runtime telemetry (e.g., power, temperature, utilization) and hardware counters (e.g., instruction-level parallelism, cache miss rates) to construct predictive and adaptive reliability estimators. As opposed to the traditional models, our AI-based system advances its state dynamically with online learning methods of training, allowing its internal state to identify fault and self-optimize it. Findings suggest that the suggested AI system can make accurate predictions in both system performance degradation and reliability estimates with a success level of 96.8 and 94.2 per cent respectively when operating in a heterogeneous environment. Comparative research with classical models of reliability, i.e. Markov and Weibull-based models reveal high gains on adaptability, precision and generalization. Moreover, the optimization using reinforcement learning provided a 1525 percentage ratio in task scheduling performance expressed in a limited thermal and power budget. This article adds to the increasing overlap between system modeling based on AI and the optimization of heterogeneous computing, sets the stage of the introduction of new intelligent reliability management of data centers, autonomous computing systems, and HPC infrastructure

The paper presents an electromyographic based (EMG based) human-machine interface system. To retain a constraint-free user environment, EMG sensing is limited to three arm muscles. EMG signals are processed to attain parameters that are related to the muscles' temporal activities. Using these parameters, a unique signature is constructed for each particular gesture. The problem of gesture classification is carried out in a framework of statistical pattern recognition. Experimental investigation was carried out to examine the system's reliability in recognizing 12 arm gestures. The results show that the system can recognize the 12 gestures with a success rate of 96%.

Size, complexity, and human dependency on software--based products have grown dramatically during past decades. Software developers are struggling to deliver reliable software with acceptable level of quality, within given budget and schedule. One measure of software quality and reliability is the number of residual faults. Therefore, researchers are focusing on the identification of the number of fault presents in the software or identification of program modules that are most likely to contain faults. A lot of models have been developed using various techniques. A common approach is followed for software reliability prediction utilizing failure data. Software reliability and quality prediction is highly desired by the stakeholders, developers, managers, and end users. Detecting software faults early during development will definitely improve the reliability and quality in cost-effective way.KeywordsSoftware ReliabilityFault PredictionSoftware MetricsCapability Maturity ModelTest Case PrioritizationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Accurate and dynamic reliability modeling for the running manufacturing system is the prerequisite to implement preventive maintenance. However, existing studies could not output the reliability value in real time because their abandonment of the quality inspection data originated in the operation process of manufacturing system. Therefore, this paper presents an approach to model the manufacturing system reliability dynamically based on their operation data of process quality and output data of product reliability. Firstly, on the basis of importance explanation of the quality variations in manufacturing process as the linkage for the manufacturing system reliability and product inherent reliability, the RQR chain which could represent the relationships between them is put forward, and the product qualified probability is proposed to quantify the impacts of quality variation in manufacturing process on the reliability of manufacturing system further. Secondly, the impact of qualified probability on the product inherent reliability is expounded, and the modeling approach of manufacturing system reliability based on the qualified probability is presented. Thirdly, the preventive maintenance optimization strategy for manufacturing system driven by the loss of manufacturing quality variation is proposed. Finally, the validity of the proposed approach is verified by the reliability analysis and optimization example of engine cover manufacturing system.

Purpose. Development of the reliability calculating methodology of fuzzy systems with components, reliability parameters of which are described by the membership functions of various kinds. Methodology. Systematic analysis of the known approaches to the assessment of the reliability of fuzzy systems with vague boundaries and different kinds of membership functions. Findings. A unified description and the new tabular algorithm of various arithmetic operations run between the various types of fuzzy sets are proposed and tabular methods of reliability analysis of the basic structures of fuzzy systems are developed. Originality. A generalized description of triangular and trapezoid-shaped accessory functions of fuzzy boundaries in a special tuples is introduced, analytical expressions for the formation of tabular schedules of their functions and the calculation of the reliability of fuzzy systems with complex structures are obtained: sequential, redundant - parallel, serial-parallel, parallel-serial. The practical value.The obtained models and tabular algorithms simplify the calculations of the reliability of complex electro-mechanical human-machine automation systems with fuzzy parameters and vague boundaries.

Security is one of the key challenges for the development of distributed collaborative manufacturing systems. Most of the access control models of collaborative systems cannot prevent illegal accesses to column/row levels of a database table, which could be caused by command injection. This paper firstly presents the possible injection attacks in RDBMS-based collaborative system on the Semantic Web and gives their classification. Based on this classification, the paper then describes an attempt at detecting command injections to the distributed collaborative manufacturing systems by comparing the inputs related nodes that are intended with the resulting parse tree in run time. To validate the proposed approach, a prototype system called “SemGuard” has been developed. By testing on two commonly used RDF databases and one real Semantic Web application, it shows that SemGuard is efficient, taking less than 10 ms overhead to protect the context information of a distributed collaborative manufacturing system.

As the most attractive realization of component-based software, composite Web service asks a heavy demand for reliability prediction in the early stage for service discovery and service composition. Although it is believed that input should be one of the most important aspects affecting the reliability of a component-based system, research of both giving more details on an operational profile and considering software architecture to get more accurate reliability prediction is scant. This paper describes an algebra-based reliability prediction approach in three aspects: (i) a sub-set of OWL-S that provides reliability estimation information is discussed; (ii) a framework that describes syntax and reliability prediction semantics of the sub-set is explained and implemented in Maude; (iii) an example of how to use the framework to calculate reliability is studied.

Scenarios are a popular means for capturing behavioural requirements of software systems early in the lifecycle. Scenarios show how components interact to provide system level functionality. If component reliability information is available, scenarios can be used to perform early system reliability assessment. In this paper we present a novel automated approach for predicting software system reliability. The approach involves extending a scenario specification to model (1) the probability of component failure, and (2) scenario transition probabilities derived from an operational profile of the system. From the extended scenario specification, probabilistic behaviour models are synthesized for each component and are then composed in parallel into a model for the system. Finally, a user-oriented reliability model described by Cheung is used to compute a reliability prediction from the system behaviour model. The contribution of this paper is a reliability prediction technique that takes into account the component structure exhibited in the scenarios and the concurrent nature of component-based systems. We also show how implied scenarios induced by the component structure and system behaviour described in the scenarios can be used to evolve the reliability prediction.

Plug-in Hybrid Electric Vehicle (PHEV) have been widely used today. Grid to Vehicle (G2V) and Vehicle to Grid (V2G) will be the main challenge of the future smart networks with the increasing population of electric vehicles (EVs). Apart from all the advantages of massive use of electric vehicles, this can cause some problems in the electricity network system, such as the reliability and power quality of the distribution network system. For instance, when the battery charging takes place, the transformer may become overloaded, and another problem such as faster aging of a transformer may occur. Utilization of PV in charging station (SPKLU) or parking lot (PVP) of electric vehicles may become a solution. Another possible solution is to separate the general load or customers from the electric vehicle loads through additional utility transformers, so that the reliability and power quality of general customers will not be impacted by the electric car battery charging activity. In this research, the combined effects of EVs and PVP on the loading and aging of utility transformers are examined based on IEEE Std C57.91-1995. EVs load profile is modeled with monte carlo simulations and tested on a distribution system with 4 different scenarios. It was obtained that the use of PVP and the separation of EVs and general loads may improve the reliability and quality of the distribution network system and maintain transformer lifetime.

High product quality and reliability are critical in the medical device industry. Accurate reliability prediction during the product development stage provides inputs for the design strategy and boosts understanding and confidence in product reliability before products are released to the market. A Bayesian framework provides a straightforward solution for product system reliability prediction with quantified uncertainty. This is achieved by propogating the uncertainty of model parameters, use conditions, and component-level reliability to the system level. In cases where a single data source is insufficient to accurately estimate reliability due to sample size limitations or potential biases, Bayesian methods are flexible to aggregate various sources of data to reduce uncertainty. In this work, Bayesian models are applied in the Design for Reliability process to assess medical device system reliability with confidence, where frequentist methods may be approximate or non-existent.