With rising environmental concerns and fossil fuel depletion, electric vehicles (EVs) have emerged as a promising alternative for sustainable transportation; however, selecting the most suitable EV remains challenging due to multiple conflicting criteria and uncertainty in expert evaluations. Since criterion weights can significantly influence the final decision, relying solely on subjective weighting may introduce bias due to differences in expert preferences and judgments. Therefore, along with subjective weighting, an objective weighting mechanism is needed to derive the importance of criteria. To address this issue, an entropy measure is used to determine objective criterion weights, while SWARA-based subjective weights and the integrated weights are considered to examine the effect of different weighting perspectives on EV ranking through q-ROFS-COPRAS. The entropy-derived weights are further compared with existing q-ROFS entropy measures and employed in q-ROFS-TOPSIS, q-ROFS-VIKOR, q-ROFS-WASPAS, q-ROFS-ARAS, and q-ROFS-MABAC to examine the consistency of the resulting rankings. The q-ROFS-COPRAS method yields the ranking E_2 ≻ E_4 ≻ E_6 ≻ E_1 ≻ E_3 ≻ E_5 , which is identical to the rankings obtained using q-ROFS-TOPSIS, q-ROFS-MABAC, and q-ROFS-WASPAS, indicating high degree of consistency among these methods. Furthermore, the applicability of the entropy-based framework is demonstrated through an existing EV selection case study, where the obtained results show substantial agreement with those reported in the literature. Sensitivity analysis for q ranging from 2 to 10 yields an unchanged final ranking, indicating the stability of the decision outcome with respect to variations in the q-rung parameter.
In recent years, metaheuristic optimization algorithms have emerged as effective tools for identifying optimal parameter values. This paper presents an effective parameter identification method based on a newly proposed optimizer, namely the Modified Stochastic Paint Optimizer (mSPO). This algorithm is applied to identify the parameters of the Shepherd’s electric vehicle model. In this work, the original Stochastic Paint Optimizer (SPO) is enhanced by incorporating a memory-based elite solution guidance strategy to produce the mSPO. This modification helps to avoid the common pitfalls in high-dimensional optimization landscapes, such as premature convergence, and prevents the algorithm from becoming trapped in local minima. Additionally, the adaptive mutation strategy is employed to dynamically adjust the balance between exploration and exploitation based on the results of previous iterations. These improvements enable the mSPO to provide a faster and more accurate convergence curve, making it a highly effective tool for solving complex optimization problems such as battery parameter identification. The performance of the mSPO is compared with six other competitive algorithms, including the SPO, COOT, ROA, AOA, OOA, and PSO. The comparison encompasses optimizing ten different CEC2022 benchmark functions as well as the Li-ion battery models. Based on the average cost function, the mSPO is ranked first among the other optimizers. Furthermore, the identification method using the modified mSPO proved highly effective in ascertaining the battery’s unidentified characteristics. The lowest average cost function value of 0.01028 was attained by mSPO, followed by ROA at 0.01223, whereas the worst value of 0.07409 was obtained by AOA. Therefore, the outstanding results of the mSPO proved that it is a promising competitor.
The stochastic Nesterov accelerated gradient (SNAG) method is widely used in large-scale machine learning. It computes the stochastic gradient at a look-ahead point generated by the momentum history. The performance of SNAG is sensitive to the choice of step size, and a suitable step size can improve its numerical convergence behavior. Therefore, designing an effective step-size strategy is important for stochastic accelerated optimization. This paper focuses on the SNAG framework and proposes an adaptive step-size scheme. Specifically, we introduce a stochastic Nesterov accelerated gradient method with an adaptive Polyak step size, termed SNAG-APS. SNAG-APS evaluates both the stochastic loss and gradient at the Nesterov extrapolation point and uses a momentum-dependent scaling factor in the adaptive step size. The main contributions of this paper include: (i) We design SNAG-APS by combining a Polyak-type adaptive step size with stochastic Nesterov acceleration. (ii) Under standard smoothness assumptions and the weak growth condition, we establish a nonconvex stationarity bound for SNAG-APS, which contains a decreasing term and a residual term. (iii) Experiments on classical machine learning datasets and deep neural network benchmarks show that SNAG-APS achieves competitive numerical performance compared with several optimization algorithms.
Robust evaluation of healthcare services is essential for safeguarding the quality of care, expanding access, and allocating resources judiciously. It also aids in identifying strengths and weaknesses within healthcare services, enabling informed decisions to enhance overall patient care. To meet this need, a confidence level-integrated group decision-making framework is developed. The expert evaluations are modelled using disc spherical fuzzy numbers (DSFNs), an advanced extension of fuzzy theory known for its ability to effectively capture multidimensional uncertainty. This study introduces two novel Yager norm-based aggregation operators for DSFNs, which are embedded in the decision-making model. The weights of experts are determined using the Rank Order Centroid (ROC) method, while the subjective and objective criteria weights are determined using the Analytical Hierarchy Process (AHP) and Criteria Importance Through Intercriteria Correlation (CRITIC) method, respectively. The confidence levels of experts are included in the decision-making to ensure practical relevance. Lastly, the healthcare services are ranked using the Complex Proportional Assessment (COPRAS) method. The results indicate that communication, expertise, and equipment availability are of the highest importance. Among the hospitals evaluated, Hospital B is found to perform the best. The comparative analysis establishes the practicality of the model, while the sensitivity analysis assesses its robustness. The resulting model provides a transparent, versatile decision-support tool that can sharpen managerial insight and strengthen evidence-based policy in healthcare systems.
This paper presents a machine learning framework for the automatic classification of rodent navigation strategies in the Barnes maze using geometric feature statistics extracted from trajectory data. Five geometric descriptors are computed from each trajectory. Five statistical measures are then calculated for each descriptor, resulting in a 25-feature representation. A classification methodology based on the Linear Discriminant Analysis (LDA) algorithm is proposed, with experimental results demonstrating adequate performance across a range of classification metrics with a mean accuracy of 83.68
