A unique adaptive neuro fuzzy inference system for optimum decision making process in a natural gas transmission unit

Abstract

In this study, a unique adaptive neuro fuzzy inference system for optimization of decision making process in natural gas transmission unit is presented. To do this, macro-ergonomics and integrated resilience engineering factors are considered as outputs to assess operators' performance and decision styles. Evaluation of decision-making styles of control room operators would help managers adjust job specification with human characteristics. In this regard, a pertinent standard questionnaire is designed to collect required data. Operators' decision styles are identified by standard questionnaire and, then, their efficiency values are calculated by considering macro-ergonomics factors through a unique adaptive neuro-fuzzy inference system (ANFIS). Moreover, fuzzy data envelopment analysis (FDEA) model is applied to validate the obtained results. Analysis of variance is used to investigate the results of ANFIS. The results show that the best decision style is flexible DM style wherein information is pertinently used as needed and there are multiple focuses for making decisions. In addition, the results reveal that information flow, safety, system efficiency, redesign, preparedness, and learning have the lowest efficiency values amongst macro-ergonomics and integrated resilience engineering factors and require more attention. Then, DM speed and violation of regulations obtain the best results in the gas transmission unit. This is the first study that introduces a unique intelligent adaptive

AIS Data-based Decision Model for Navigation Risk in Sea Areas

“…It is usually difficult to determine the supremum and the infimum of objective eigenvalues when making decisions by the optimal absolute membership degree model, resulting in subjective arbitrariness of decision making. Calculation of the objective optimal relative membership degree and the decision-making optimal relative membership degree by using the fuzzy optimisation model can effectively avoid the subjectivity of the optimal absolute membership degree model (Azadeh et al, 2016; Pan et al, 2016). In addition, the physical meaning of the model is clear, the theory of the model is rigorous and the calculation of the model is simple (Hu and Xu, 2013).…”

Stability analysis of brain emotional intelligent controller with application to electrically driven robot manipulators

“…'Fault-tolerance, parallelism and excellent learning capabilities are other beneficial characteristics of fuzzy systems and neural networks' [11,12]. These outstanding properties have been the main motivations for widespread applications of neuro-fuzzy systems in different fields such as non-linear, robust and adaptive control [13][14][15][16][17], time-series prediction [18,19], decision making [20][21][22] and robotics [23][24][25].…”

Direct adaptive model-free control of a class of uncertain nonlinear systems using Legendre polynomials

“…Fault-tolerance, parallelism and excellent learning capabilities are other beneficial characteristics of fuzzy systems and neural networks (Khorashadizadeh and Fateh, 2017). These outstanding properties have been the main motivations for widespread applications of neuro-fuzzy systems in different fields such as nonlinear, robust and adaptive control (Hsu, 2013; Kundu and Parhi, 2017; Orlowska-Kowalska et al, 2010, Rao et al, 2017; Salahshour et al, 2018, Khorashadizadeh and Sadeghijaleh, 2018; Zaidi et al, 2017), signal processing (Engin, 2004; Güler and Übeyli, 2005), time-series prediction (Miranian and Abdollahzade, 2013; Nayak et al, 2004), decision making (Azadeh et al, 2016; Zheng et al, 2015) and robotics (Petković et al, 2012, 2016; Van Pham and Wang, 2015; Zhou et al, 2015). In neuro-fuzzy control, two general approaches can be distinguished: direct (Hsueh et al, 2014) and indirect (Li et al, 2014).…”