•This study proposes an integrated approach for identifying hotspot locations on a roadway network.•The method assists transportation authorities to prioritize hotspot locations more efficiently and ...allocate their limited budget and resources.•The method allows to analyse the spatial coverage of crashes.•The Network KDE is improved using the exposure data.
This paper proposes a two-step integrated method for identifying traffic accident (TA) hotspots on a roadway network. The first step includes a spatial analysis method called network kernel density estimation (KDE). The second step is a network screening method using the critical crash rate, which it described in the Highway Safety Manual (HSM). The method was examined by using three years of TAs (2011–2013) in Sherbrooke, Canada. The network KDE uses TAs to graphically display sites with a high crash density. Two different crash patterns were used for identifying these locations: (1) a crash pattern that includes three-year aggregated crash data, and (2) a crash pattern that involves three-year merged crash data. The results of the two crash patterns were evaluated based on a prediction accuracy index (PAI). It was found that the results obtained from the merged crash data outperformed the other. On the other hand, crash clustering in a site does not imply a site is hotspot and it is better to tested by other factors. High crash density locations were then tested by the critical crash rate, which helps to create an accurate comparison of sites. The importance of the critical crash rate is that it takes several factors into account such as the amount of exposure, the type of intersection, variance in crash data, etc. We realized that the hotspots determined using the two methods reflect very problematic locations and filter out the locations that do not have a problem. This approach could help transportation authorities and safety specialists to identify and prioritize sites that require more safety attention.
This article presents a novel data-based approach to investigate the non-Gaussian stochastic distribution control problem. As the motivation of this article, the existing methods have been summarised ...regarding to the drawbacks, for example, neural network weights training for unknown stochastic distribution and so on. To overcome these disadvantages, a new transformation for dynamic probability density function is given by kernel density estimation using interpolation. Based upon this transformation, a representative model has been developed while the stochastic distribution control problem has been transformed into an optimization problem. Then, data-based direct optimization and identification-based indirect optimization have been proposed. In addition, the convergences of the presented algorithms are analysed and the effectiveness of these algorithms has been evaluated by numerical examples. In summary, the contributions of this article are as follows: 1) a new data-based probability density function transformation is given; 2) the optimization algorithms are given based on the presented model; and 3) a new research framework is demonstrated as the potential extensions to the existing stochastic distribution control.
In this study, we tried to investigate the possible correlation of inter and intra-band quadrupole transition probabilities of rotational bands. To this aim, all the available experimental values of ...quadrupole transition rates between different levels of the ground band together quadrupole transitions of gamma and beta bands, 2γ+→0g+ and 2β+→0g+, of even-even prolate nuclei in the 150<A<250 mass region are analyzed. We extended the kernel density estimation approach as a non-parametric estimation formalism for this analysis and compared its predictions with the results of the commonly used parametric technique with emphasis on Porter-Thomas distribution. The results show that the statistical correlation is dominant for such inter-band transitions between different levels of the ground band but for intra-band transitions which originated from the beta band, deviation from correlated behavior is obvious. Also, the comparison of the statistical behavior of 2+→0+ transitions of different rotational bands, suggests the most GOE-like behavior for the quadrupole transitions between the levels of the ground band.
The shipping industry has long been hindered by piracy, which threatens the lives of the crews of ships as well as maritime security. Understanding the characteristics of maritime piracy across ...different scales can help optimize global routes to reduce shipping risks and improve maritime security. In this study, we propose a framework for exploring and comparing the characteristics of maritime piracy at global and regional scales. Data on piracy incidents were collected from Global Integrated Shipping Information System, developed by the International Maritime Organization, and were reconstructed by using text mining and geospatial techniques. The distribution of the key variables and the spatiotemporal distribution of incidents of maritime piracy were investigated, and records of these incidents were divided into six categories. The differences in the characteristics of maritime piracy across categories were then quantitatively examined. The results show that the Gulf of Guinea had the largest number of pirates, and the distance from incidents to the shoreline, the weapons used by pirates, the types and status of the attacked ships, and the area that the incidents frequented varied significantly across regions. The results of this study can be used by the shipping industry to reduce the risk of maritime piracy.
This article concerns the issue of data-driven fault diagnosis for series lithium-ion battery pack. A voltage correlation-based statistical analysis method is proposed. First, the voltage of each ...cell within the battery pack is measured independently, and the correlation coefficient (CC) between the voltages of adjacent cells is calculated. Then, all the CC signals under normal conditions are used to train a principal component analysis model, on the basis of which synthetic statistic and kernel density estimation-based thresholds are designed for simultaneously monitoring all the measured CC signals at each sampling instant. Once a fault is detected, an accumulative relative contribution plot algorithm is immediately used to isolate which CC signal has a problem and locate the faulty cell through a cross-positioning strategy. The experimental results on a realistic test platform for series battery pack show that the new method provides accurate and reliable assessments for different fault specifics, and it performs better than the state-of-the-art CC methods in terms of parallel processing capability, sensitivity to weak short circuit faults, and robustness to window width.
There is a growing interest among scientists about climate change and its adverse effects. One of the major adverse effects of climate change is the sea level rise (SLR), which will cause habitat ...loss for many species and threaten their survival. Sea turtles are an example of animal groups most likely to be threatened by SLR. It is, therefore, critical to predict the effect of SLR on sea turtle habitats to prepare better conservation and management plans that consider the climate change impact. With this aim, we projected the outcomes of SLR on the habitat and nest loss of one of the most important Mediterranean green sea turtle (Chelonia mydas) nesting beaches (Samandag, Turkey) using natural nests between 2008 and 2016 nesting seasons. Under the extreme scenario (1.2 m SLR) one-third of the coastal area and up to 18% of natural nests could be lost at a key green turtle nesting beach for this globally unique population.
•One of the major negative impacts of climate change is sea level rise (SLR)•We predicted the impacts of SLR scenarios on the green turtle (Chelonia mydas) habitat and nest loss•One-third of the coastal area and nearly one-fifth of the natural nests could be lost•The SLR due to climate change would have negative effect on the reproductive success of Samandağ green turtle population
Uncertainty of getting admission into universities / institutions is one of the global challenges in academic environment. The students are having good marks with high credential but not sure about ...getting their admission into universities / institutions. In this research study the researcher built a predictive model using Naïve Bayes Classifiers –machine learning algorithm to extract and analyze hidden pattern in students’ academic records and their credentials. The main objective of this research study is to reduce uncertainty for getting admission into universities / institutions on the basis of their previous credentials and some other essentials parameters. This research study presents a joint venture of Naïve Bayes Classification and Kernel Density Estimations (KDE) approach to predict student’s admission into universities or any higher institutions. The predictive model is built on training dataset of students’ examination score such as GPA, GRE, RANK and some other essential features that offered the admission with a predictor accuracy rate of 72% and has been experimentally verified. To improve the quality of accuracy of predictive model the researcher used the Shapiro-Walk Normality Test and Gaussian distribution on large datasets. The predictive model helps in reducing the admission uncertainty and enhances the universities decision making capabilities. The significance of this research study is to reduce human intervention for making decisions with respect to students’ admission into universities or any higher academic institutions, and it demonstrates that many universities and higher-level institutions could use this predictive model to improve their admission process without human intervention.
Effective prediction of water demand is a prerequisite for decision makers to achieve reliable management of water supply. Currently, the research on water demand prediction focuses on point ...prediction method. In this study, we constructed a GA-BP-KDE hybrid interval water demand prediction model by combining non-parametric estimation and point prediction. Multiple metaheuristic algorithms were used to optimize the Back-Propagation Neural Network (BP) and Kernel Extreme Learning Machine (KELM) network structures. The performance of the water demand point prediction models was compared by the Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Kling-Gupta Efficiency (KGE), computation time, and fitness convergence curves. The kernel density estimation method (KDE) and the normal distribution method were used to fit the distribution of errors. The probability density function with the best fitting degree was selected based on the index G. The shortest confidence interval under 95% confidence was calculated according to the asymmetry of the error distribution. We predicted the impact indicator values for 2025 using the exponential smoothing method, and obtained water demand prediction intervals for various water use sectors. The results showed that the GA-BP model was the optimal model as it exhibited the highest computational efficiency, algorithmic stability, and prediction accuracy. The three prediction intervals estimated after adjusting the KDE bandwidth parameter covered most of the sample points in the test set. The prediction intervals of the four water use sectors were evaluated as F values of 1.6845, 1.3294, 1.6237, and 1.3600, which indicates high accuracy and quality of the prediction intervals. The mixed water demand interval prediction based on GA-BP-KDE reduces the uncertainty of the point prediction results and can provide a basis for water resource management by decision makers.
•Using the GA-BP-KDE model to quantify the uncertainty in water demand forecasting.•Metaheuristics were adopted for hyperparameter tuning of BP and KELM.•The interval estimation sample set was divided into three tiers.•The shortest 95% confidence intervals of errors of point prediction were obtained.•Interval water demand of 21 areas in Sichuan for 2025 was predicted.
PDF
