Short-term Forecast for Average Speed of Road Section based on Floating Car Data with Support Vector Regression

The particleboard (PB) production is an extremely complex process, many operating parameters affecting panel quality. It is a big challenge to optimize the PB production parameters. The production parameters of particle gluing have an important influence on the internal bond (IB) strength of PB. In this study, using grey relation analysis (GRA) and support vector regression (SVR) algorithm, a prediction model was developed to accurately predict IB of PB through particle gluing processing parameters in a PB production line. GRA was used to analyze the grey relational grade between the particle gluing processing parameters and IB of PB, and the variables were screened. The SVR algorithm was used to train 724 groups of particle gluing sample data between six particle gluing processing parameters and IB. The SVR model was tested with 181 sets of experimental data. The SVR model was verified by 181 sets of experimental data, and the values of mean absolute error (MAE), mean relative error (MRE), root mean square error (RMSE), and Theil’s inequality coefficient (TIC) of the model were 0.008, 0.017, 0.013, and 0.014, respectively. The results showed that the prediction performance of the nonlinear regression prediction model based on GRA–SVR is superior, and the GRA–SVR prediction model can be used to real-time predict the IB in the PB production line.

Traffic data are the fundamental inputs to traffic flow analysis and simulation studies, which facilitate decision making in the fi eld of traffic engineering. Hence, the accuracy of traffic data is of paramount importance. This study compares new technologies available for traffic data collection considering their accuracy and applicability in the Sri Lankan context. Traffic in Sri Lanka is of heterogeneous nature, as opposed to the homogeneous nature observed in most developed countries. Hence, collection of traffic data poses several challenges that affects its accuracy. Three techniques, the infrared driven TIRTL instrument, the video image processing-based TRAZER application and the traffic data collection method using the Google distance matrix application programming interface (API), with respect to their data collection accuracy are reviewed in this study. The fundamental macroscopic traffic data variables (fl ow and speed) were evaluated against control surveys. It was found that each technology has its strengths and weaknesses and needs to be used appropriately. The TIRTL instrument fared better on road sections on level terrain when the crossfall did not obstruct the infrared beams. Such occasions provided a rich set of microscopic traffic data. The TRAZER software delivered data up to a 100 % accuracy. However, this required the user to go through a lengthy postprocessing routine to extract the final set of traffic data. Google traffic data collection provides highly accurate results when estimating link speeds. This method is ideal for collection of bulk data with spatio-temporal variations and the process can be fully automated to reduce the human resource requirement.

To ensure the availability and quality of drugs, Public Health Centers (PHC) must pay attention to the planning and procurement process. The problem that often arises is the increase in drug stock due to the stable use of drugs each month, resulting in excess and expired drugs that are not used. In addition, it is necessary to avoid inappropriate drug demand, which affects stock availability. Drug usage prediction is done with several methods such as the Single Moving Average (SMA) algorithm in the data mining method and the Support Vector Regression (SVR) algorithm in the machine learning method. This algorithm was chosen because the drug data of Diazepam 5 mg and Mefenamic Acid 500 mg is sustainable from January 2020 to June 2023 (42 months). Implementation using the Phyton programming language. Testing using the Mean Absolute Percentage Error (MAPE) method, this study aims to measure the accuracy of predictions in each algorithm. In research with Diazepam 5 mg and Mefenamic Acid 500 mg drugs, with a division of 80% in training data and 20% in test data. With a calculation of 3 periods, the SMA algorithm produces MAPE values of 4.10% and 4.29%, in the "very good" range. The SVR algorithm, which uses an RBF kernel with a complexity parameter of 1.0 and an epsilon parameter of 0.1, produces MAPE results of 7.35% and 9.52%, in the "Very Good" range. Thus, the SMA algorithm predicts better than the SVR algorithm.

In order to test the applicability of support vector regression (SVR) in near-infrared (NIR) spectral analysis of tobacco,NIR calibration model for starch in tobacco were developed by processing NIR spectra and chemically determined starch content data of 187 tobacco samples with SVR,partial least square regression (PLS),multiplicative linear regression (MLR) and error back propagation artificial neural network (BP-ANN). The obtained model was tested through internal validation and external validation with leave-one-out cross validation (LOOCV) and independent sample set. The results showed that the accuracy of SVR model was slightly higher than that of BP-ANN,PLS and MLR models,it implied that SVR algorithm could be applied to NIR analysis of starch in tobacco.

In order to solve the shortcoming of support vector regression algorithm, sports economic activities income forecasting based on genetic support vector regression algorithm is presented in this paper. As the kernel parameter, insensitive loss parameter and penalty parameter have a great influence on the forecasting performance of support vector regression algorithm, genetic algorithm is used to perform the parameters optimization of support vector regression algorithm simultaneously in this study. The experimental results show that the sports economic activities income forecasting results of genetic support vector regression model trained by the four-dimension training samples are more excellent than those of genetic support vector regression model trained by the three-dimension training samples, and the sports economic activities income forecasting ability of genetic support vector regression model trained by the four- dimension training samples is better than that of support vector regression model trained by the four-dimension training samples. Thus, genetic support vector regression algorithm has a good application prospect in sports economic activities income forecasting.

Accurate forecast of carbon dioxide (CO2) emissions plays a significant role in China's carbon peaking and carbon neutrality policies. A novel two-stage forecast procedure based on support vector regression (SVR), random forest (RF), ridge regression (Ridge), and artificial neural network (ANN) is proposed and evaluated by comparing it with the single-stage forecast procedure. Nine independent variables’ data (study period: 1985–2020) are used to forecast the CO2 emissions in China. Our results reveal that, when the time gap, h increases from 1 to 8, the average root mean squared error (RMSE) and mean absolute error (MAE) of SVR–SVR, SVR–RF, SVR–Ridge, and SVR–ANN are almost uniformly lower than errors arising from their single-stage version, respectively. Among these two-stage models, SVR–ANN exhibits the lowest forecast errors, whereas SVR–RF admits the highest. The mean percentage decrease in forecast errors of SVR–SVR vs. SVR, SVR–RF vs. RF, SVR–Ridge vs. Ridge, and SVR–ANN vs. ANN are 36.06, 5.98, 43.05, and 14.81 for RMSE, and 36.06, 6.91, 43.27, and 15.35 for MAE. Our two-stage procedure is also suitable to forecast other variables, such as fossil fuel and renewable energy consumption.

Accurate prediction of Rate of Penetration (ROP) is critical for optimizing drilling efficiency and reducing costs in hydrocarbon exploration. Traditional ROP models often overlook formation temperature, despite its significant influence on rock mechanics and drilling fluid rheology, particularly in high-temperature sandstone reservoirs like those in the Niger Delta. This study employs three machine learning (ML) algorithms—Support Vector Regression (SVR), Random Forest (RF), and Artificial Neural Networks (ANN)—to evaluate the contribution of formation temperature to ROP prediction. A dataset of 1,200 drilling records from Niger Delta wells was used, incorporating parameters such as weight-on-bit (WOB), rotary speed (RPM), pump pressure, and formation temperature. Model performance was assessed using R², Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), with ablation studies to isolate temperature’s impact. Inclusion of formation temperature improved ROP prediction accuracy across all models. The ANN achieved the highest performance (R² = 89.3%, RMSE = 0.387, MAE = 0.141), followed by RF (R² = 90.5%, RMSE = 2.737, MAE = 0.900) and SVR (R² = 87.8%, RMSE = 0.553, MAE = 0.169). Temperature omission led to significant performance degradation (R² reductions of 7–13%). Sensitivity analysis ranked temperature among the top three influential features. Formation temperature is a critical but underutilized parameter in ROP modeling. ML techniques, particularly ANN, demonstrate superior capability in capturing nonlinear temperature-dependent effects, offering actionable insights for real-time drilling optimization in thermally complex formations. This study provides a framework for integrating thermal data into predictive models to enhance drilling efficiency in the Niger Delta and analogous basins.

Risk premium plays an important role in stock investing. Experiments have shown that value stocks typically have a higher average return than growth stocks; however, this effect persists indefinitely, even disappearing in some stages. Some studies suggested high volatility in the series of returns, broken structures, market volatility, or the impact of financial crises. This study aimed to build the uncertainty index and control it in the regression analysis model to solve the limitations above. The empirical analysis in Ho Chi Minh Stock Exchange (HOSE) showed that a value premium exists, and value stocks have a higher average return than growth stocks due to the higher overall risk. Furthermore, this study combined the Support Vector Regression (SVR) algorithm with the risk premium theoretical framework for the forecasting model; consequently, it is the most efficient model.

To improve the accuracy of short-term traffic state prediction, this paper proposes a short-term traffic state prediction method based on the fusion of Gravitational search algorithm (GSA) and Support Vector Regression (SVR) to predict the average speed and traffic on urban roads. The average speed is predicted. This method combines the working mechanism of SVR in prediction and uses the GSA algorithm to optimize the relevant parameters in the SVR kernel function to improve the accuracy of the SVR kernel function, thereby obtaining more accurate prediction accuracy. Taking 4 road sections in a road network in Xuanwu District of Nanjing City as an example, using the measured traffic flow and speed data of urban roads to test the performance of the GSA-SVR model. Using 12 historical average data from 17:00-17:55 are recorded as training sample to predict road traffic flow and speed data from 18:00-18:25, the analysis results show that from the overall prediction rate, the GSA-SVR algorithm reduces the prediction error to less than 40% of the SVM algorithm, which verifies the GSA-SVR mentioned in this article. The effectiveness and practicality of the algorithm can provide a reliable basis for decision-making for the good management of urban roads.

Fracture toughness evaluation of ground granulated blast furnace slag concrete using experimental study and machine learning techniques

Floating car data (FCD) is an important source of traffic data for the design and management of an intelligent transportation system, and the efficiency and accuracy of FCD map-matching influences the application of FCD. Most map-matching algorithms are very time-consuming because many links need to be evaluated as to whether they are the correct link for the travelling floating car, especially in urban areas where most FCD are collected. In fact, the highest degree of map-matching a link for an FCD is not always the correct link, so the use of candidate links is proposed here. In this study, candidate links are defined as those that fall within the FCD error. We propose a novel algorithm for quickly identifying accurate candidate links on the basis of a vector to raster road map conversion approach. First, a road buffer is constructed based on the size of the confidence region by taking into account the position error of the FCD and the width of the link. Second, all link buffers are converted to a raster map with resolution in metres, each pixel position of the raster map corresponding to a geographical coordinate. A pixel can have one or more values, which represent the link ID whether the link buffer covers the pixel; so a spatial index map between geographical coordinate and link ID is built in this way. Then, candidate links are calculated rapidly in the next three steps: (1) Selecting a link ID from the spatial index map that, according to the FCD positioning fix coordinate, must fall within the confidence region of the FCD. The number is very small and this step is very fast because it does not include a traverse process. (2) Getting data for selected link from the road networks vector map by the link ID. This step uses a binary search and is very fast. (3) Calculating the degree of map-matching of these links to identify candidate links of the FCD. This algorithm is far less time-consuming than earlier approaches. The algorithm was tested using more than 13 million real-world FCD collected in Wuhan City, China, and the performance was shown to be satisfactory.

The purpose of this study was to present a new strategy based on fusion at the decision level for modeling the crop residue. To this end, a set of satellite imagery and field data, including the Residue Cover Fraction (RCF) of corn, wheat and soybean was used. Firstly, the efficiency of Random Forest Regression (RFR), Support Vector Regression (SVR), Artificial Neural Networks (ANN) and Partial-Least-Squares Regression (PLSR) in RCF modeling was evaluated. Furthermore, to increase the accuracy of RCF modeling, different algorithms results were combined based on their modeling error, which is called the decision-based fusion strategy. The R2 (RMSE) between the actual and modeled RCF based on ANN, RFR, SVR and PLSR algorithms for corn were 0.83 (3.89), 0.86 (3.25), 0.76 (4.56) and 0.75 (4.81%), respectively. These values were 0.81 (4.86), 0.85 (4.22), 0.78 (5.45) and 0.74 (6.20%) for wheat and 0.81 (3.96), 0.83 (3.38), 0.76 (5.01) and 0.72 (5.65%) for soybean, respectively. The error of corn, wheat and soybean RCF estimating decision-based fusion strategy was reduced by 0.90, 0.96 and 0.99%, respectively. The results showed that by implementing the decision-based fusion strategy, the accuracy of the RCF modeling was significantly improved.

Significant wave height estimation using SVR algorithms and shadowing information from simulated and real measured X-band radar images of the sea surface

Commercial floating car data application in Sub-Saharan African transport planning contexts: A critical review and research agenda

Use of automated learning techniques for predicting mandibular morphology in skeletal class I, II and III