Search results

The purpose of this paper is to propose an improved optimization methodology of information granulation‐based fuzzy radial basis function neural networks (IG‐FRBFNN). In the IG‐FRBFNN, the membership functions of the premise part of fuzzy rules are determined by means of fuzzy c‐means (FCM) clustering. Also, high‐order polynomial is considered as the consequent part of fuzzy rules which represent input‐output relation characteristic of sub‐space and weighted least squares learning is used to estimate the coefficients of polynomial. Since the performance of IG‐RBFNN is affected by some parameters such as a specific subset of input variables, the fuzzification coefficient of FCM, the number of rules and the order of polynomial of consequent part of fuzzy rules, we need the structural as well as parametric optimization of the network. The proposed model is demonstrated with the use of two kinds of examples such as nonlinear function approximation problem and Mackey‐Glass time‐series data.

The type of polynomial of each fuzzy rule is determined by selection algorithm by considering the local error as performance index. In addition, the combined local error is introduced as a performance index considered by two kinds of parameters such as the polynomial type of each rule and the number of polynomial coefficients of each rule. Besides this, other structural and parametric factors of the IG‐FRBFNN are optimized to minimize the global error of model by means of the hierarchical fair competition‐based parallel genetic algorithm.

The performance of the proposed model is illustrated with the aid of two examples. The proposed optimization method leads to an accurate and highly interpretable fuzzy model.

The proposed hybrid optimization methodology is interesting for designing an accurate and highly interpretable fuzzy model. Hybrid optimization algorithm comes in the form of the combination of the combined local error and the global error.

This article presents a systematic framework with a meta-analytic approach to finding various antecedents, consequents and moderating effects of trust in financial services.

A meta-analysis of 165 articles was performed, which generated 272 observations in a cumulative sample of 86,968 respondents.

The results of this meta-analysis demonstrated seventeen antecedents of trust constructs and four consequents. Most of these relationships were meaningful and consistent. The authors also found some significant moderators related to culture (individualism, masculinity and long-term orientation) and context (innovation index and device type).

This meta-analysis reviewed the relationships found throughout the theoretical framework about the trust construct in financial service contexts, identifying new paths for future research. Some limitations, such as the non-use of qualitative studies and the selection of concepts, exist in the secondary data and should be noted.

The present study can assist financial system managers in decision-making because the findings from the meta-analysis are more consistent than those from traditional primary surveys.

This research tested the impact of antecedents, consequents and moderators of trust in the financial services sector and presented significant results using a meta-analytic review. This meta-analysis contributes to the marketing literature by offering a set of empirical generalizations, including relationship coefficients and fail-safe calculated numbers (FSN).

The purpose of this paper is to consider the concept of Fuzzy Radial Basis Function Neural Networks with Information Granulation (IG‐FRBFNN) and their optimization realized by means of the Multiobjective Particle Swarm Optimization (MOPSO).

In fuzzy modeling, complexity, interpretability (or simplicity) as well as accuracy of the obtained model are essential design criteria. Since the performance of the IG‐RBFNN model is directly affected by some parameters, such as the fuzzification coefficient used in the FCM, the number of rules and the orders of the polynomials in the consequent parts of the rules, the authors carry out both structural as well as parametric optimization of the network. A multi‐objective Particle Swarm Optimization using Crowding Distance (MOPSO‐CD) as well as O/WLS learning‐based optimization are exploited to carry out the structural and parametric optimization of the model, respectively, while the optimization is of multiobjective character as it is aimed at the simultaneous minimization of complexity and maximization of accuracy.

The performance of the proposed model is illustrated with the aid of three examples. The proposed optimization method leads to an accurate and highly interpretable fuzzy model.

A MOPSO‐CD as well as O/WLS learning‐based optimization are exploited, respectively, to carry out the structural and parametric optimization of the model. As a result, the proposed methodology is interesting for designing an accurate and highly interpretable fuzzy model.

This study aims to propose, as the first time, the interval type-2 adaptive network-fuzzy inference system (ANFIS) structure, which is given better results compared to previously presented in the open literature. So, the ANFIS can be used effectively for training of interval type-2 fuzzy logic system (IT2FLS) parameters.

Karnik–Mendel algorithm (KMA) is modified to use in interval type-2 ANFIS. The modified Karnik–Mendel algorithm (M-KMA) is implemented to change the uncertain ANFIS parameters into known ones. In this way, the interval type-2 ANFIS removes uncertainties of IT2FLS. Therefore, the interval type-2 ANFIS is reduced to a simple one, i.e. less mathematical operation required. Only consequent parameters are trained, and the consequent parameters are chosen in the form of crisp.

By applying the mentioned procedure, it can be shown that interval type-2 ANFIS has generally better results compared to type-1 ANFIS. However, it was noticed that the worst results obtained in the case of interval type-2 ANFIS are equal to the best result obtained in the case of type-1 ANFIS. Therefore, users in this field can use this approach in solving nonlinear problems.

The interval type-2 ANFIS can be used as controller for highly nonlinear systems such as air vehicles.

As stated in the open literature, it is ineffective to use ANFIS for IT2FLS. In this study, the KMA is modified for IT2FLS, and it is seen that the ANFIS can be used effectively for IT2FLS.

This paper presents an application of the fuzzy gradual rules in an epidemic study of canine rabies in São Paulo city, Brazil. A linguistic epidemiological model was elaborated through fuzzy rules built by the Extension Principle. We used both the inference method of Mamdani and of Dubois et al. The results were compared with real data from São Paulo and with another MISO Model, which is entirely based on expert knowledge presented in a previous work. Questions about application of fuzzy techniques in epidemiology, different inference methods and the Dubois et al. methodology are discussed.

The purpose of this paper is to analyze electromagnetic performance and develop an analytical approach to find the suitable coil combination and no-load flux linkage of the proposed hybrid excited consequent pole flux switching machine (HECPFSM) while minimizing the drive storage and computational time which is the main problem in finite element analysis (FEA) tools.

First, a new HECPFSM based on conventional consequent pole flux switching permanent machine (FSPM) is proposed, and lumped parameter magnetic network model (LPMNM) is developed for the initial analysis like coil combination and no-load flux linkage. In LPMNM, all the parts of one-third machine are modeled which helps in reduction of drive storage, computational complexity and computational time without affecting the accuracy. Second, self and mutual inductance are calculated in the stator, and dq-axis inductance is calculated using park transformation in the rotor of the proposed machine. Furthermore, on-load performance analysis, like average torque, torque density and efficiency, is done by FEA.

The developed LPMNM is validated by FEA via JMAG v. 19.1. The results obtained show good agreement with an accuracy of 96.89%.

The proposed HECPFSM is developed for high-speed brushless AC applications like electric vehicle (EV)/hybrid electric vehicle (HEV).

The proposed HECPFSM offers better flux regulation capability with enhanced electromagnetic performance as compared to conventional consequent pole FSPM. Moreover, the developed LPMNM reduces drive storage and computational time by modeling one-third of the machine.

The purpose of this paper is to investigate an alternative simplified analytical approach for the design of electric machines. Numerical-based finite element method (FEM) is a powerful tool for accurate modelling and electromagnetic performance analysis of electric machines. However, computational complexity, magnetic saturation, complex stator structure and time consumption compel researchers to adopt alternate analytical model for initial design of electric machine especially flux switching machines (FSMs).

In this paper, simplified lumped parameter magnetic equivalent circuit (LPMEC) model is presented for newly developed segmented PM consequent pole flux switching machine (SPMCPFSM). LPMEC model accounts influence of all machine parts for quarter of machine which helps to reduce computational complexity, computational time and drive storage without affecting overall accuracy. Furthermore, inductance calculation is performed in the rotor and stator frame of reference for accurate estimation of the self-inductance, mutual inductance and dq-axis inductance profile using park transformation.

The developed LPMEC model is validated with corresponding FEA using JMAG Commercial FEA Package v. 18.1 which shows good agreement with accuracy of ∼98.23%, and park transformation precisely estimates the inductance profile in rotor and stator frame of reference.

The model is developed for high-speed brushless AC applications.

The proposed SPMCPFSM enhance electromagnetic performance owing to partitioned PMs configuration which make it different than conventional designs. Moreover, the developed LPMEC model reduces computational time by solving quarter of machine.

This paper aims to reviewed analytical methodologies, i.e. lumped parameter magnetic equivalent circuit (LPMEC), magnetic co-energy (MCE), Laplace equations (LE), Maxwell stress tensor (MST) method and sub-domain modelling for design of segmented PM(SPM) consequent pole flux switching machine (SPMCPFSM). Electric machines, especially flux switching machines (FSMs), are accurately modeled using numerical-based finite element analysis (FEA) tools; however, despite of expensive hardware setup, repeated iterative process, complex stator design and permanent magnet (PM) non-linear behavior increases computational time and complexity.

This paper reviews various alternate analytical methodologies for electromagnetic performance calculation. In above-mentioned analytical methodologies, no-load phase flux linkage is performed using LPMEC, magnetic co-energy for cogging torque, LE for magnetic flux density (MFD) components, i.e. radial and tangential and MST for instantaneous torque. Sub-domain model solves electromagnetic performance, i.e. MFD and torque behaviour.

The reviewed analytical methodologies are validated with globally accepted FEA using JMAG Commercial FEA Package v. 18.1 which shows good agreement with accuracy. In comparison of analytical methodologies, analysis reveals that sub-domain model not only get rid of multiples techniques for validation purpose but also provide better results by accounting influence of all machine parts which helps to reduce computational complexity, computational time and drive storage with overall accuracy of ∼99%. Furthermore, authors are confident to recommend sub-domain model for initial design stage of SPMCPFSM when higher accuracy and low computational cost are primal requirements.

The model is developed for high-speed brushless AC applications.

The SPMCPFSM enhances electromagnetic performance owing to segmented PMs configuration which makes it different than conventional designs. Moreover, developed analytical methodologies for SPMCPFSM reduce computational time compared with that of FEA.

The purpose of this paper is to propose an alternative approach to approximate reasoning by DNA computing, thereby adding a new dimension to the existing approximate reasoning method by bringing it down to nanoscale computing. The logical aspect of approximate reasoning is replaced by DNA chemistry.

To achieve this goal, first the synthetic DNA sequence fuzzified by quantum dot, which is a recent advancement of nanotechnology. Thus with the help of fuzzy DNA, which holds the vague concept of human reasoning, the basic method of approximate reasoning on a DNA chip is realized. This approach avoids the tedious choice of a suitable implication operator (for a particular application) necessary for existing approximate reasoning based on fuzzy logic. The inferred consequences obtained from DNA computing‐based approximate reasoning is ultimately hybridized with appropriate complementary sequence probed on a DNA‐chip to confirm the result of inference.

The present approach is suitable for reasoning under vague and uncertain environment and does not require any subject choice of any individual expert, which is essential for existing approximate reasoning method.

This new tool for approximate reasoning based on DNA computing is applicable to several problems of science and engineering; namely pattern classification, control theory, weather forecasting, atmospheric science, etc.

The purpose of this paper is to investigate whether and how knowledge hoarding, functions as antecedent and consequent of work related negative acts, as a measure of bullying. The authors investigate the relation as mediated by trust and justice.

Data stem from a longitudinal study in which questionnaire responses were collected twice from 1,650 employees in 52 workplaces. Structural equation modelling was used to analyse the two models. Design-based corrections were made to accommodate the multi-level structure of data.

The analyses showed that knowledge hoarding was both an antecedent and a consequent of negative acts. First, over time, knowledge hoarding was indirectly related to negative acts mediated by trust and justice. Second, negative acts were both directly and indirectly related to knowledge hoarding over time. The study thus points to the existence of a vicious circle of negative acts, psychological states of trust and justice, and knowledge hoarding behaviours, which presumably will affect both individual and organizational outcomes negatively.

The use of already collected, self-report data, single-item measures, and the two-year time lag could pose potential limitations to the study.

Preventive and repair actions could potentially impact both negative acts and knowledge hoarding by focusing on increasing the social exchange quality at work unit level.

This paper combines two strands of research, that of bullying at work and that of knowledge management, within which research on knowledge hoarding has been an under-researched area.