Search

: Non-linear neutrosophic numbers (NLNNs) are different kinds of neutrosophic numbers with at least one non-linear membership function (either of truthiness, falsity or indeterminacy part) of the information. Furthermore, a linear programming problem with non-linear neutrosophic numbers as coefficients/parameters is a special type of programming problem known as a non-linear linear programming problem (NLN-LPP). This paper elaborates on the concepts of non-linear neutrosophic number (NLNN) sets, different forms of non-linear neutrosophic numbers (NLNNs),α, β, γ cuts on non-linear neutrosophic numbers (NLNNs), possibility mean, possibility standard deviation, and possibility variance of non-linear neutrosophic numbers (NLNNs). In this paper, we also propose the solution technique for non-linear neutrosophic linear programming problems (NLN-LPPs) in which all coefficients/parameters are non-linear neutrosophic numbers (NLNNs). In this continuation, we suggest a new modified possibility score function for non-linear NNs in terms of possibility means and possibility standard deviations of non-linear neutrosophic numbers (NLNNs) for better use of all parts of information. This modified score function is used to convert non-linear neutrosophic number (NLNN) coefficients/parameters of non-linear neutrosophic linear programming problem (NLN-LPP) into equivalent crisp values. Thereafter, the equivalent crisp problem is solved with the usual method to obtain the optimal solution of non-linear neutrosophic linear programming problem (NLN-LPP). The proposed solution algorithm is unique and new for solving non-linear neutrosophic linear programming problems. A numerical example is solved with the proposed algorithm to legitimate the research output. A case study is also discussed to show its applicability in solving real-life problems. [ABSTRACT FROM AUTHOR]

Due to high costs associated with conventional maintenance strategies, application of soft computing in monitoring the condition of equipment to predict the health of various components of machine tools in manufacturing processes has attracted the attention of researchers. Soft computing is a better alternate to predict and optimize the manufacturing processes related to physics-based models as these processes are complex and precarious. The theories of artificial neural systems, fuzzy logic, genetic algorithms, ant colony optimization, simulated annealing and particle swarm optimization are utilized by soft computing techniques to handle real-world issues that cannot be palatably handled utilizing conventional computing methods. This paper presents a state-of-the-art review on the recent developments in the use of soft computing in condition-based maintenance in manufacturing. [ABSTRACT FROM AUTHOR]