Ensemble-based ship weather multi-objective route optimization

Application of a deep learning-based discrete weather data continuousization model in ship route optimization

Bayesian optimization based extreme gradient boosting and GPR time-frequency features for the recognition of moisture damage in asphalt pavement

Estimation of internal states in a Li-ion battery using BiLSTM with Bayesian hyperparameter optimization

In recent decades, various software effort estimation (SEE) algorithms have been suggested. Unfortunately, generating high-precision accuracy is still a major challenge in the context of SEE. The use of traditional techniques and parametric approaches is largely inaccurate because they produce biased and subjective accuracy. Meanwhile, none of the machine learning methods performed well. This study applies the AdaBoost ensemble learning method and random forest (RF), on the other hand the Bayesian optimization method is applied to determine the hyperparameters of this model. The PROMISE repository and the ISBSG dataset were used to build the SEE model. The developed model was comprehensively compared with four machine learning methods (classification and regression tree, k-nearest neighbor, multilayer perceptron, and support vector regression) under 3-fold cross validation (CV). It can be seen that the RF method based on AdaBoost ensemble learning and bayesian optimization outperforms this approach. In addition, the AdaBoost-based model assigns a feature importance rating, which makes it a promising tool in software effort prediction.

Data driven strength and strain enhancement model for FRP confined concrete using Bayesian optimization

In this paper, we study two classes of optimal reinsurance models by minimizing the total risk exposure of an insurer under the criteria of value at risk (VaR) and conditional value at risk (CVaR). We assume that the reinsurance premium is calculated according to the expected value principle. Explicit solutions for the optimal reinsurance policies are derived over ceded loss functions with increasing degrees of generality. More precisely, we establish formally that under the VaR minimization model, (i) the stop-loss reinsurance is optimal among the class of increasing convex ceded loss functions; (ii) when the constraints on both ceded and retained loss functions are relaxed to increasing functions, the stop-loss reinsurance with an upper limit is shown to be optimal; (iii) and finally under the set of general increasing and left-continuous retained loss functions, the truncated stop-loss reinsurance is shown to be optimal. In contrast, under CVaR risk measure, the stop-loss reinsurance is shown to be always optimal. These results suggest that the VaR-based reinsurance models are sensitive with respect to the constraints imposed on both ceded and retained loss functions while the corresponding CVaR-based reinsurance models are quite robust.

Optimization of metabolizable energy prediction models for maize in laying hens by incorporating anti-nutritional factors

This study compares emissions and transit times from an environmentally oriented and a time oriented routing of large freight shipments in the European rail/road transportation network. We use the terminal-and-service selection problem (TSSP) to find the optimal routings under the different objectives. We show that substantial differences exist between the emission oriented routing and the time oriented routing. A large-scale simulation study reveals that shipments in the emission minimizing routing emit on average almost half as much emissions as if they were routed with the objective to minimize transit time. At the same time, the average transit time of shipments in the emission oriented routing almost triples compared to the transit time in the time optimal routing. This shows by experiment that substantial emission reductions can be achieved in the European freight transport sector by a corresponding routing of shipments but that this comes at the cost of a much lower service quality.

Road crashes cause significant traffic delay, which can bring unnecessary financial losses. The objective of this study is to predict the level of delay caused by crashes (LDC) and discuss significant risk factors. To ensure the efficiency and accuracy of prediction, an improved stacking model was developed using Texas crash data of 2020. The first layer integrates seven base classifiers and the second layer tests three classifiers with different advantages. To improve and simplify the stacking model, three state-of-the-art methods—Bayesian hyperparameter optimization (BO), multiobjective feature selection (FS), and ensemble selection (ES)—were used. First, the hyperparameters and the least and most effective features were selected for each base classifier by BO and FS, respectively. Then ES, considering diversity and performance, selects the least base classifiers to reduce the input of the second layer. Finally, permutation feature importance was used to interpret the best stacking model. The results indicate that the stacking model achieves superior performance on four indicators: recall, G mean, F1 score, and area under the receiver operating characteristic (ROC) curve (AUC-ROC). FS significantly improves the efficiency of the stacking model and ES obtains a simplified stacking model without significantly reducing performance. In addition, the combination of the two methods (FS and ES) tends to achieve the best performance, and six risk factors have the greatest contributions in prediction using permutation feature importance. The prediction of LDC and the analysis of the main contributing factors help road managers respond to the rescue strategies to mitigate traffic congestion caused by crashes in a timely manner, thus minimizing economic losses.

Calculation of the mean long-term service speed of transport ship: Part III Influence of shipping route and ship parameters on its service speed Service speed obtainable by a ship in real weather conditions when sailing on a given shipping route, is one of the major parameters which have great impact on ship operation costs. The so far used, very approximate method of service speed prediction based on "service margin", is very little exact. In this paper a new method based on additional ship resistance dependent on mean statistical parameters of wave and wind occurring on a given shipping route, is presented. The mean long-term service speed is calculated on the basis of the calculated additional resistance and the screw propeller and propulsion engine parameters. Also, a new definition of service margin and a way of its calculation is presented apart from the results of the mean service speed calculation depending on ship's type and size and shipping route.

Diamond grinding wheels (DGWs) have a central role in cutting-edge industries such as aeronautics or defense and spatial applications. Characterizations of DGWs are essential to optimize the design and machining performance of such cutting tools. Thus, the critical issue of DGW characterization lies in the detection of diamond grits. However, the traditional diamond detection methods rely on manual operations on DGW images. These methods are time-consuming, error-prone and inaccurate. In addition, the manual detection of diamond grits remains challenging even for a subject expert. To overcome these shortcomings, we introduce a deep learning approach for automatic diamond grit segmentation. Due to our small dataset of 153 images, the proposed approach leverages transfer learning techniques with pre-trained ResNet34 as an encoder of U-Net CNN architecture. Moreover, with more than 8600 hyperparameter combinations in our model, manually finding the best configuration is impossible. That is why we use a Bayesian optimization algorithm using Hyperband early stopping mechanisms to automatically explore the search space and find the best hyperparameter values. Moreover, considering our small dataset, we obtain overall satisfactory performance with over 53% IoU and 69% F1-score. Finally, this work provides a first step toward diamond grinding wheel characterization by using a data-driven approach for automatic semantic segmentation of diamond grits.

Maritime transportation is essential for global trade but presents significant environmental challenges due to its greenhouse gas emissions. Existing studies have addressed these challenges through integrated routing and speed optimization frameworks, yet frequently lack explicit quantification of environmental impacts and exhibit limited scalability for large-scale ship routing operations. Conversely, existing quantum optimization research in vehicle routing predominantly targets land-based transportation scenarios, restricting its direct applicability to maritime logistics. Maritime logistics inherently involve distinct operational complexities, such as nonlinear interactions among speed, payload, fuel consumption, and numerous operational uncertainties. These combined limitations underscore the critical need for quantum optimization methods explicitly designed for green maritime supply chains. To bridge this gap, this paper proposes an efficient quantum-centric optimization framework that uses the quantum approximate optimization algorithm (QAOA) to jointly optimize ship routing and speed management within sustainable maritime supply chains. Specifically, we formulate an NP-hard cost minimization problem integrating critical maritime parameters, including fuel consumption, payload constraints, and operational speeds. We further develop a hybrid quantum-classical alternating optimization approach that iteratively addresses routing decisions through quantum computing techniques and optimizes ship speed using an analytical solution. Simulation results and real quantum hardware experiments demonstrate that our quantum-centric methodology achieves substantial cost reductions and highlights the potential for practical applicability in realistic maritime operations, significantly outperforming classical optimization benchmarks.

The paper presents new statistical approaches for modeling highly variable mechanical properties and screening specimens in development of new materials. Particularly, for steels, Charpy V-Notch (CVN) exhibits substantial scatter which complicates prediction of impact toughness. The paper proposes to use Conditional Value-at-Risk (CVaR) for screening specimens with respect to CVN. Two approaches to estimation of CVaR are discussed. The first approach is based on linear regression coming from the Mixed-Quantile Quadrangle, and the second approach builds CVN distribution with percentile regression, and then directly calculates CVaR. The accuracy of estimated CVaR is assessed with some variant of the coefficient of multiple determination. We estimated discrepancy between estimates derived by two approaches with the Mean Absolute Percentage error. We compared VaR and CVaR risk measures in the screening process. We proposed a modified procedure for ranking specimens, which takes into account the uncertainty in estimates of CVaR.

The COVID-19 pandemic first broke out in 2020 and rapidly spread globally, which had a great impact on shipping companies, ports and cargo owners. Many nations have closed their ports and raised quarantine requirements, leading to frequent incidents of ship skipping and ship stoppages. Although the COVID-19 pandemic has been brought under control, similar or more serious public health events may still occur in the future worldwide, which will have a greater impact on ship transportation. This study focuses on optimizing the route and speed of container ships in the pandemic environment by constructing a mixed integer programming model, which provides better response strategies for ship operators. Existing studies have reported that the pandemic has significantly impacted international ships’ ocean-going routes, resulting in an increase in the total cost of ship operations on the routes. Moreover, ships that on ocean going routes were less affected by the pandemic, and generally can cope well with the pandemic after speed optimization; while ships on near-ocean routes are more susceptible to the pandemic, and are more likely to optimize costs by choosing a port-hopping strategy. The study also found that the choice of ship routes is affected by factors (e.g., the number of containers loaded and unloaded, the distance between the strictly controlled port and the next port, and the range of port quarantine period). Fuel prices have no significant impact on the choice of ship routes.

This study will solve the classical vehicle routing problem, the goal is to generate k trips with the shortest distance for h customers with predetermined locations and needs.. The proposed solution to the classical vehicle routing problem is a hybrid sine cosine algorithm. The sine cosine algorithm is hybridized with the grey wolf optimizer, which is used in combination with the methods of tournament selection, opposition learning, and the mutation and crossover method to build the optimal routing plan for the means of transporting cement. To demonstrate the advantages of the developed hybrid sine cosine algorithm, this algorithm is evaluated and compared with modern algorithms such as sine cosine algorithm, dragonfly algorithm, grey wolf optimizer, ant lion optimizer, particle swarm optimization, modified hybrid particle swarm optimization, genetic algorithm, and the double-population genetic algorithm in case studies. The hybrid sine cos algorithm gives optimal results in these cases because it balances mining and exploration. Thus, the results of this study indicate that managers can use the developed hybrid sine cosine algorithm to create optimal vehicle routing plans to reduce transportation distances.