Soft Computing, Год журнала: 2022, Номер 27(7), С. 3829 - 3845
Опубликована: Ноя. 3, 2022
Язык: Английский
Mathematics, Год журнала: 2023, Номер 11(6), С. 1454 - 1454
Опубликована: Март 17, 2023
Seaport network efficiency is very crucial for global maritime economic trades and growth. In this work, data of three years (2018–2020) with input variables (time in port, age vessels, size cargo carrying capacity vessels) output (Liner Shipping Connectivity Index (LSCI) Gross Domestic Product (GDP)) are collected. Few screening tests performed to ensure the fit further analyses. Since none existing studies has ever considered LSCI as an variable, main purpose study measure seaport based on using envelopment analysis (DEA), both classical fuzzy. fuzzy DEA, triangular number (TrFN) trapezoidal (TpFN) used construct sets scores DEA. The comparison between DEA (TrFDEA) shows range differences results ranges from −0.0274 0.0105, while (TpFDEA) yields within −0.0307 0.0106. Using relative reference, it revealed that TpFDEA smaller standard deviations variances than TrFDEA 2018 2019, whereas opposites hold true during pandemic year 2020. With use numbers, uncertainty levels measurement can be investigated minimum, mean, median maximum values taken into consideration. Moreover, proposed lead new insights boundedness concept scores, which were never reported before by any researcher, especially industry research. Fuzzy regression modelling Possibilistic Linear Regression Least Squares (PLRLS) method was also determine interval minimum connectivity efficiencies, gave a better estimation regular model.
Язык: Английский
Процитировано
8
Proceedings of the Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science, Год журнала: 2023, Номер 238(5), С. 1573 - 1589
Опубликована: Июль 18, 2023
The exploration of the prediction wired electro-discharge machining parameters is becoming progressively more crucial in order to choose ideal for making products with impeccable designs. objective this work develop an augmented adaptive neuro-fuzzy inference system (ANFIS) model by encrypting snake optimizer estimating surface roughness, material removal rate, and residual stress Monel 400 alloy wire electric discharge (WEDM). performance indicated called ANFIS-SO was evaluated using 64 trials WEDM experimental data. As a means emphasizing presented model, estimation results were compared those from existing models, which included integration ANFIS beetle antennae search capabilities, reptile algorithms, COOT algorithms. showed excellent based on statistical benchmarks tested developed models. achieved RMSE 0.0793, 0.2781, 16.3286 errors 0.3203, 1.9687, 1112.529 along accuracy 99%, 98%, 95% MRR, stress, respectively, are extremely accurate than ANFIS, ANFIS-BAS, ANFIS-RSA, ANFIS-COOT. result these findings, it evident that might be used reliability when designing future forecasting approaches optimizing parameters. Therefore, discovered innovative method determining very promising.
Язык: Английский
Процитировано
7
Processes, Год журнала: 2024, Номер 12(3), С. 479 - 479
Опубликована: Фев. 27, 2024
In today’s industrial landscape, the imperative of fault warning for equipment and systems underscores its critical significance in research. The deployment not only facilitates early detection identification potential failures, minimizing downtime maintenance costs, but also bolsters reliability safety. However, intricacies non-linearity inherent data often pose challenges to traditional methods, resulting diminished performance, especially with complex datasets. To address this challenge, we introduce a pioneering approach that integrates an enhanced Coati Optimization Algorithm (ICOA) Bidirectional Long Short-Term Memory (Bi-LSTM) network. Our strategy involves triple incorporating chaos mapping, Gaussian walk, random walk mitigate randomness initial solution conventional (COA). We augment search capabilities through dual population strategy, adaptive factors, stochastic differential variation strategy. ICOA is employed optimal selection Bi-LSTM parameters, effectively accomplishing prediction task. method harnesses global COA sophisticated analysis enhance accuracy efficiency warnings. practical application real-world case induced draft fan warning, our results indicate anticipates faults approximately two hours advance. Furthermore, comparison other advanced namely, Improved Social Engineering Optimizer Optimized Backpropagation Network (ISEO-BP), Sparrow Particle Swarm Hybrid Light Gradient Boosting Machine (SSAPSO-LightGBM), Butterfly (MSBOA-Bi-LSTM), proposed exhibits distinct advantages robust effects.
Язык: Английский
Процитировано
2
Buildings, Год журнала: 2024, Номер 14(6), С. 1713 - 1713
Опубликована: Июнь 7, 2024
Musculoskeletal disorders, which are epidemiologically related to exposure whole-body vibration (WBV), frequently self-reported by workers in the construction sector. Several activities during building and demolition expose this physical agent. Directive 2002/44/CE defined a method of assessing WBV that was limited an eight-hour working day, did not consider cumulative long-term effects on health drivers. This study aims propose methodology for generating individualised models vehicle drivers exposed easy implement companies, ensure is compromised short or long term. A measurement campaign conducted with professional driver, collected data were used formulate six artificial neural networks predict daily compressive dose lumbar spine assess short- exposure. Accurate results obtained from developed network models, R2 values above 0.90 training, cross-validation, testing. The approach proposed offers new tool can be applied assessment workers’ their life subsequent retirement.
Язык: Английский
Процитировано
2
Soft Computing, Год журнала: 2022, Номер 27(7), С. 3829 - 3845
Опубликована: Ноя. 3, 2022
Язык: Английский
Процитировано
10