Summary Marine predator algorithm (MPA) is a powerful metaheuristic optimization algorithm that shows effective convergence ability on complex benchmark functions. The combination of Brownian and Levy flight distributions directly affects the convergence strategy of MPA. Although MPA has good convergence performance, it is open to improvement as it falls to a local optimum and cannot comprehensively scan the search area during the exploration phase. In this study, MPA has been improved by integrating elite natural evolution and elite random mutation strategies. In addition, these two strategies are combined with Gaussian mutation. The proposed method in this study which is named as elite evolution strategy MPA (EEMPA) has achieved comprehensive scanning of the solution space and considerably reduced the risk of falling into the local optimum trap, with elite strategies. The effect of EEMPA has been tested with the CEC2017 and CEC2019 benchmark functions. EEMPA has been compared with some metaheuristic algorithms frequently used in the literature and gives promising results among the considered optimization methods. Furthermore, EEMPA has been examined for seven well‐known real world engineering problems. When the results are compared with both classical MPA and enhanced MPA methods, EEMPA converges to better than the other methods.

Abstract In the present article, the efficiency of a new hybrid learning method, named artificial neural network with gradient‐based optimizer algorithm (ANN‐GBO), is investigated to determine a local geoid. The outcomes of the assessed method are compared with classical ANN (without GBO), some metaheuristic‐based ANN models and other study results (interpolation methods). Four commonly used performance metrics, root mean square error (RMSE), mean absolute error (MAE), mean absolute relative error (MARE) and coefficient of determination () have been used to assess the applied methods. Assessment of predictions revealed that ANN‐GBO yielded the best results (RMSE: 10.51 cm, MAE: 8.26 cm, MARE: 0.27 cm, and : 0.9622). The outcomes of the work clearly show that the ANN‐GBO approach has the lowest prediction error compared with ANN. It can be said that the GBO algorithm enhances the ANN capability in local geoid modeling. It may be recommended to optimize the weights of the ANN with GBO for high prediction accuracy. This research can be extended to other regions.

In this paper, based on power aggregation (PA) operators, neutrality average and neutrality geometric aggregation operators are proposed. Furthermore, a general score function for q-rung orthopair fuzzy sets (q-ROFSs) is proposed. PA can reduce the impact of excessively high or excessively low arguments. It also emphasizes the interrelationship between attributes on the decision matrix. PA and neutrality aggregations (NAs) were hybrid with more fair and neutral decisions of the DMs evaluations. Also, neutrality aggregation operator provides reliable results by taking into account neutrality among decision-makers. On the one hand, through its dynamic structure, q-ROFSs provide a wider evaluation for decision-makers (DMs). q-ROFSs include many fuzzy sets with varying q ≥ 1 parameters. One of the most important parts of multi-attribute group decision-making problems is that DMs do not take into account their bias. On the other hand, the NA operator evaluates the decisions of the DMs from a neutral attitude perspective. The power neutrality aggregation operator proposed in this study produces more objective results. The proposed operator is more consistent with the advantages of both operators and it deals with the attitude of DMs more objectively. Validity tests are applied to the proposed methods. Also, the superior aspects of the proposed methods are demonstrated by numerical examples. As a result of validity tests and comparisons with other studies in the literature, it is seen that the proposed methods give effective results.