Predicting queue wait time probabilities for multi-scale computing

A generalized data preprocessing method is proposed in this paper to reduce the amount of outliers among historical data and further improve the power prediction accuracy. Historical data of wind farms are fit with an S-shape curve via Linear Regression Model. Based on this statistical curve, outliers can be identified considering different fitting error. Furthermore, the expansion of wind farm is identified through the number of outliers. Then a selection method for the allowed maximum fitting errors is recommended. The presented method has been integrated into the prediction system in Inner Mongolia of China with 36 farms. The actual application shows that the wind farm power prediction accuracy has been improved by at least 28% with this model. It is noteworthy that the proposed preprocessing method is just based on statistical analysis of historical data and thus compatible with various wind power prediction methods.

Force generation by actin assembly shapes cellular membranes. An experimentally constrained multiscale model shows that a minimal branched actin network is sufficient to internalize endocytic pits against membrane tension. Around 200 activated Arp2/3 complexes are required for robust internalization. A newly developed molecule-counting method determined that ~200 Arp2/3 complexes assemble at sites of clathrin-mediated endocytosis in human cells. Simulations predict that actin self-organizes into a radial branched array with growing ends oriented toward the base of the pit. Long actin filaments bend between attachment sites in the coat and the base of the pit. Elastic energy stored in bent filaments, whose presence was confirmed by cryo-electron tomography, contributes to endocytic internalization. Elevated membrane tension directs more growing filaments toward the base of the pit, increasing actin nucleation and bending for increased force production. Thus, spatially constrained actin filament assembly utilizes an adaptive mechanism enabling endocytosis under varying physical constraints.

In this poster contribution, we present a scheduling system for automated remote operation of instruments at high-altitude research facilities and similar remote sites. Via web-based interfaces, the system allows instrument owners as well as authorized third-party scientists to schedule and execute measurements and observations.The system has been developed as a thesis project in the context of the AlpEnDAC-II ("Alpine Environmental Data Analysis Centre", www.alpendac.eu) collaboration (funded by the Bavarian State Ministry of the Environment and Consumer Protection). Consequently, the scheduler and interfaces have been integrated with the AlpEnDAC Operating-on-Demand functionalities. A first use case for the framework has been the operation of an airglow imager (FAIM) in Oberpfaffenhofen (DE).We describe the design and implementation of our system for scheduling and execution of multi-user observations on instruments, including scheduling-data transfers and data retrieval. Our core implementation uses an optimization-based scheduler (Google's OR-Tools) to ensure maximum instrument use and to minimize idle times. Results show that the scheduler is reliable, fast, and is consistently able to provide optimal observation plans. The extensibility of the system is guaranteed by the usage of modern software in the core of the system, including well-defined and specified communication through REST APIs. Thus, it can easily be adapted to other settings and instruments, which is also facilitated by a modern deployment strategy using Docker and Kubernetes.

This paper describes the development of an artificial intelligence (AI) system for survival prediction from intraocular melanoma. The system used artificial neural networks (ANNs) with five input parameters: coronal and sagittal tumour location, anterior tumour margin, largest basal tumour diameter and the cell type. After excluding records with missing data, 2331 patients were included in the study. These were split randomly into training and test sets. Date censorship was applied to the records to deal with patients who were lost to follow-up and patients who died from general causes. Bayes theorem was then applied to the ANN output to construct survival probability curves. A validation set with 34 patients unseen to both training and test sets was used to compare the AI system with Cox's regression (CR) and Kaplan–Meier (KM) analyses. Results showed large differences in the mean 5 year survival probability figures when the number of records with matching characteristics was small. However, as the number of matches increased to >100 the system tended to agree with CR and KM. The validation set was also used to compare the system with a clinical expert in predicting time to metastatic death. The rms error was 3.7 years for the system and 4.3 years for the clinical expert for 15 years survival. For <10 years survival, these figures were 2.7 and 4.2, respectively. We concluded that the AI system can match if not better the clinical expert's prediction. There were significant differences with CR and KM analyses when the number of records was small, but it was not known which model is more accurate.

The subject of this study is modern architectural approaches to the development of Single Page Applications (SPA), specifically a comparative analysis of classical client-side rendering (SPA), server-side rendering (Server-Side Rendering, SSR), and static site generation (Static Site Generation, SSG). The research focuses on key aspects that determine the choice of architecture for real projects: performance during content loading and rendering, as well as support for search engine optimization (SEO). As practical comparison objects, two widely used technologies were selected: Create React App, representing a pure SPA approach, and the framework Next.js, which implements SSR and SSG models. All examined frontend options are built on a unified backend – a REST API on Node.js, ensuring the correctness of the comparison under equal conditions. The methodology includes the practical implementation of three types of applications (SPA, SSR, SSG) on React/Next.js with a common REST API. Performance assessment was conducted using Google Lighthouse based on FCP, LCP, and Speed Index metrics. The analysis of SEO accessibility was performed through cURL requests simulating the behavior of search engine bots Googlebot and Yandexbot. The scientific novelty of the work lies in the comprehensive comparative experiment carried out, which quantitatively evaluates the trade-offs between three dominant architectural paradigms of web development using a unified codebase and identical API. A key conclusion is that classical SPA, despite its advantages in interactivity post-loading, falls short in critically important areas for successful applications – initial performance and SEO. This creates fundamental limitations for its use in public, content-oriented projects. The results of the study allow for the formalization of criteria for architecture selection: SSR and SSG become preferable for projects where quick initial loading and search promotion are paramount, while SPA may remain the choice for complex internal interfaces with high operational requirements. The work provides a clear practical justification for the transition from universal SPAs to hybrid or server-rendered solutions in modern frontend development.

During early mammalian embryo development, a small number of cells make robust fate decisions at particular spatial locations in a tight time window to form inner cell mass (ICM), and later epiblast (Epi) and primitive endoderm (PE). While recent single-cell transcriptomics data allows scrutinization of heterogeneity of individual cells, consistent spatial and temporal mechanisms the early embryo utilize to robustly form the Epi/PE layers from ICM remain elusive. Here we build a multiscale three-dimensional model for mammalian embryo to recapitulate the observed patterning process from zygote to late blastocyst. By integrating the spatiotemporal information reconstructed from multiple single-cell transcriptomic datasets, the data-informed modeling analysis suggests two major processes critical to the formation of Epi/PE layers: a selective cell-cell adhesion mechanism (via EphA4/EphrinB2) for fate-location coordination and a temporal attenuation mechanism of cell signaling (via Fgf). Spatial imaging data and distinct subsets of single-cell gene expression data are then used to validate the predictions. Together, our study provides a multiscale framework that incorporates single-cell gene expression datasets to analyze gene regulations, cell-cell communications, and physical interactions among cells in complex geometries at single-cell resolution, with direct application to late-stage development of embryogenesis.

We analyze the velocity anisotropy of stars in real and energy space for a sample of Milky Way–like galaxies in the TNG50 simulation. We employ different selection criteria, including spatial, kinematic, and metallicity cuts, and make three halo classes (–) that show mild-to-strong sensitivity to different selections. The above classes cover 48%, 16%, and 36% of the halos, respectively. We analyze the β radial profiles and divide them into either monotonically increasing radial profiles or ones with peaks and troughs. We demonstrate that halos with monotonically increasing β profiles are mostly from class , while those with peaks/troughs are part of classes and . This means that care must be taken, as the observationally reported peaks/troughs might be a consequence of different selection criteria. We infer the anisotropy parameter β energy space and compare that against the β radial profile. It is seen than 65% of halos with very mild sensitivity to different selections in real space are those for which the β radial and energy profiles are closely related. Consequently, we propose that comparing the β radial and energy profiles might be a novel way to examine the sensitivity to different selection criteria and thus examining the robustness of the anisotropy parameter in tracing stellar kinematics. We compare simulated β radial profiles against various observations and demonstrate that, in most cases, the model diversity is comparable with the error bars from different observations, meaning that the TNG50 models are in good overall agreement with observations.

The boiler station is a critical section in palm oil mills as it supplies steam for sterilization and energy generation. However, a case study revealed that the boiler station operated with an Overall Equipment Effectiveness (OEE) of only 27.03%, far below the 85% world-class standard. The main contributors to this inefficiency were reactive maintenance practices, delayed failure responses, and aging components with no early detection mechanisms, resulting in frequent breakdowns and prolonged downtimes. This research was conducted with three objectives: (1) to evaluate and analyze the root causes of low OEE performance, (2) to develop a predictive maintenance system using logistic regression to forecast component failures based on historical maintenance data, and (3) to integrate an automated alert system that notifies engineers of high-risk components before failures occur. Logistic regression was selected for its simplicity, interpretability, and suitability for environments with limited digital infrastructure, making it ideal for palm oil mills with structured Excel-based records. The predictive system was developed using Python and Streamlit, enabling engineers to upload maintenance logs, receive failure probability predictions, and visualize high-risk components through an interactive web interface. The model achieved a prediction accuracy of 77%, demonstrating its effectiveness in shifting maintenance practices from reactive to predictive, improving equipment reliability, and aligning with Total Productive Maintenance (TPM) and Industry 4.0 principles.

In this era of Industry 4.0, technology is advancing rapidly, and various applications are emerging to help traders optimize their product procurement. One such application is the forecasting or prediction system. Serba Serbi Online Shop Tanjungpinang is a shop that requires this system to manage its inventory successfully. As the shop offers a wide variety of goods, managing inventory has become challenging. The capital spent on inventory has increased, and unsold goods are piling up in the warehouse. To overcome this issue, the shop needs to calculate the amount of inventory required to reduce the stock build-up in the warehouse. The prediction method used for this purpose is the least square method. This method uses periodic or time-series data, which requires past sales data to predict future sales. Historical sales data is necessary to develop an inventory prediction system. The sales data for a single item from July 2022 to May 2023 was analyzed using the Least Square method. Based on this analysis, the predicted sales for June is 144.64, rounded to 145. The accuracy of the prediction results were tested, and the best accuracy was obtained from the prediction using 3 months of data, with an error rate of 5.47%. The prediction with the highest error rate was obtained from using 9 months of data, with an error rate of 10.65%. On average, the prediction accuracy is 0.02154%, which is considered very good.

In this work, we propose a Human Activity Prediction (HAP) system using activity sequence spanning trees constructed from a life-log created by a video sensor-based daily Human Activity Recognition (HAR) system using time-sequential Independent Component (IC)-based depth silhouette features with Hidden Markov Models (HMMs). In the daily HAR system, the IC features are extracted from the collection of the depth silhouettes containing various daily human activities such as walking, sitting, lying, cooking, eating etc. Using these features, HMMs are used to model the time sequential features and recognize various human activities. The depth silhouette-based human activity recognition system is used to recognize daily human activities automatically in real time, which creates a life-log of daily activity events. In this work, we propose a method for human activity prediction using fixed-length activity sequence spanning trees based on the life-log. Utilizing the consecutive activities recorded in an activity sequence database (i.e. life-log) for a specific period of time of each day over a period such as a month, the fixed-length spanning trees can be constructed for the sequences starting with each activity where the leaf nodes contain the frequency of the fixed-length consecutive activity sequences. Once the trees are constructed, to predict an activity after a sequence of activities, we traverse the spanning trees until a path up to the previous node of the leaf nodes is matched with the testing pattern. Finally, we can predict the next activity based on the highest frequency of the leaf nodes along the matched path. The prediction experiments over the computer simulated data which is based on the daily logs show satisfactory results. Our video sensor-based human activity recognition and prediction systems can be utilized for practical applications such as smart and proactive healthcare.

Along with changing technology, Human resources are still needed in many parts of decision making. The companies and organizations still use human to analysis data. Despite of that performance, human analysis took longer time and effort to complete. And sometimes there is always a negative factor of human resources such as unmanageable human. Therefore, it’s always important to provide an excellent training source so this human resource able to reach an outcome that needed. Machine learning is one of the most common knowledge that use in decision making. There are many forms of machine learning such as regression, classification, clustering, etc. two of which is used in this application, regression and classification. Naive Bayes regression is one of classification method which rooted on Bayes theorem. Naïve Bayes use historical data to predict future outcome based on the characteristic on that historical data. Multiple Linear Regression involves more than one independent variable or predictor. With machine learning and human resources, man can easily to analyse credit worthiness and determine the credit limit of one bank costumer without taking a long time and much effort.

Crime is always one of the most important social problems, and it poses a great threat to public security and people. Accurate crime prediction can help the government, police, and citizens to carry out effective crime prevention measures. In this paper, the research on crime prediction is systematically reviewed from a variety of temporal and spatial perspectives. We describe the current state of crime prediction research from four perspectives (prediction content, crime types, methods, and evaluation) and focus on the prediction methods. According to various temporal and spatial scales, temporal crime prediction is divided into short-term prediction, medium-term prediction, and long-term prediction, and spatial crime prediction is divided into micro-, meso-, and macro-level prediction. Spatio-temporal crime prediction classification can be a permutation of temporal and spatial crime prediction classifications. A variety of crime prediction methods and evaluation metrics are also summarized, and different prediction methods and models are compared and evaluated. After sorting out the literature, it was found that there are still many limitations in the current research: (i) data sparsity is difficult to deal with effectively; (ii) the practicality, interpretability, and transparency of predictive models are insufficient; (iii) the evaluation system is relatively simple; and (iv) the research on decision-making application is lacking. In this regard, the following suggestions are proposed to solve the above problems: (i) the use of transformer learning technology to deal with sparse data; (ii) the introduction of model interpretation methods, such as Shapley additive explanations (SHAPs), to improve the interpretability of the models; (iii) the establishment of a set of standard evaluation systems for crime prediction at different scales to standardize data use and evaluation metrics; and (iv) the integration of reinforcement learning to achieve more accurate prediction while promoting the transformation of the application results.

Against the backdrop of an increasingly large stock market and a growing number of investors, many individuals have faced bankruptcy due to blindly following trends in stock investment. This situation has created a demand for methods to predict stock prices in advance. To address this issue, researchers have previously proposed various approaches, including traditional analytical methods based on historical data, statistical analysis methods, machine learning and deep learning methods, etc. The research focus of this paper is to investigate the specific conditions under which the Backpropagation Neural Network model delivers superior performance in stock price prediction. This study utilizes 44 years of historical stock price data from Apple Inc. (AAPL), encompassing key features such as Date, Opening Price, Highest Price of the Day, Lowest Price of the Day, Closing Price, Adjusted Closing Price, and Trading Volume. Specifically, the study focuses on selecting the Opening Price, Highest Price of the Day, Lowest Price of the Day, Closing Price, and Trading Volume as input features. The research ultimately achieved relatively accurate results. These findings are then leveraged to extend insights into effective configuration strategies for a wider range of other prediction methods.

Commercial off-the-shelf (COTS) software has become a crucial component of many of today’s mission- and safety-critical products. As these products have proliferated internationally, global demand for software safety and performance quality has increased. Unfortunately, the current state-of-the-art in software assessment technology has not matured to the point where well-established and widely accepted schemes for evaluating COTS software quality have been presented. Without insight into the source code, COTS software users depend on a subjective assessment of attributes described by performance metrics, behavioral descriptions, and/or historical product data. Published techniques that attempt to formalize the above approach are often less than successful due to ill-defined and subjective metrics, and the naive assumption of independence between attributes. Recently, software standards have been seen as a means of improving the evaluation process by providing an acknowledged criterion for software metric comparison. Given the stated issues, is it possible to develop a standards-based dependency model of well-defined attributes that does not depend on naive assumptions? The answer to this question is yes. This paper describes a process that can be used to extract acyclic dependency models, if they exist, from various software product quality standards. Application of the extraction process will be used to reveal the attribute dependency model of the international software standard ISO/IEC 9126-1. Not only can this process be used for COTS software evaluation, but it can also be used by standards making bodies to clarify the reasoning, justifications and causal implications they employ when developing standards.

Introduction. Taking the Workers’ and Peasants’ Red Army of 1941—1945 as an example the authors identify features of the personnel selection in the army during military operations, conditions, requirements, criteria, qualities necessary for promotion to higher command positions. The aim of the study is to identify the mechanism of personnel selection in the armed forces during military operations. Methods. The authors use the following general scientific methods: modeling, structural and functional, systemic and comparative analysis; movement from the abstract to the concrete and from the concrete to the abstract. The authors make use of P.A. Sorokin’s theory of vertical mobility. For further research of the problem, the methodology of synergetics and counterfactual modeling of the past can be involved. Scientific novelty of the research. The following existing concepts of the Red Army officers’ dynamics in 1941—1945 are generalized: positive, negative, moral and psychological selection. Their basic concepts and shortcomings are revealed. An initiativeintellectual concept of the officers’ selection in war is formulated on the basis of P. Sorokin’s theory of vertical mobility. The author identify specific conditions, requirements, and criteria influencing the selection of commanders during the war. Results. A clear opposition to the conditions, requirements and selection criteria in peacetime and wartime is given. The following specific criteria for selecting officers of the Red Army during the war are highlighted: the presence of combat experience; non-conformism; the ability to take personal responsibility; analyze and correct errors; nonstandard thinking. Military action creates specific conditions that give rise to the selection criteria for commanders. Conclusions. The selection criteria for personnel in peacetime and wartime armies differ sharply. At the beginning of battle actions, stereotyped commanders prevailed. Such people are replaced by commanders who are able to go beyond the established canons, orders of the chiefs. The work presents a diagram showing the stages of selecting officers in a war: military actions — conditions — selection criteria - requirements — new qualities of a person adapted to combat operations. This leads to the success of military operations, and, consequently, to the promotion of officers in rank and position.