Real-time Decision Research Articles

Distributed Data Analysis in Cloud Services for Insurance Companies

This article embarks on an insightful journey through the realm of advanced data analysis techniques which can be used in the insurance area, with a keen focus on the applications and capabilities of Graph Neural Networks (GNN) in the following sector. The article is structured into several chapters, which include the overview of existing and commonly used approaches of the data representation, the possible ways of data analysis of the data in such a representation, deep dive into the concept of GNN for the graph data analysis and the applicability of each approach in the insurance industry. The initial chapter introduces the two main concepts of the data representation, which are the commonly used relational database and the more modern approach of dimensional data design. Then the focus is moved to the graph data representation, which also can be used for data analysis in the cloud environment. To achieve the best applicability in the insurance industry, particularly in underwriting and claims management, the article analyzes the advantages of each approach to the data representation as well as its drawbacks. To conclude the chapter, the comparison table of the three approaches is presented. Based on the comparison table, the decision to use the graph representation is made as it enables the industry to unravel complex relationships and dependencies amid various data points—such as policyholder history, incident particulars, and third-party information—resulting in more accurate risk assessments and efficient claim resolutions. Then the article presents the concept of Graph Neural Networks, a rather new concept which can be used to analyze the data, represented in a graph form using machine learning algorithms. The potential of using this approach for the data analysis in the insurance area and some possible use cases are described. The advantages of using this approach include ability to effectively capture and leverage the complex relationships inherent in graph- structured data and a powerful framework for analyzing and processing graph-structured data. However, the potential drawbacks of the approach such as complexity to design and difficulties in scaling are also considered. Further along, the article probes the strategic integration of Graph Neural Networks with real-time and dynamic data environments, examining their adaptability to evolving network patterns and temporal dependencies. We discuss how this adaptability is paramount in contexts like real-time decision-making and predictive analysis, which are crucial for staying agile in a rapidly changing market landscape. Then the exact use cases of the GNN applicability in the insurance area are provided, including the claim assignment and underwriting process are described in detail. Furthermore, the simplified mathematical formulation of the underwriting process is provided, which elaborates the role GNNs play in propelling actuarial science with their capability to incorporate node attributes, edge information, and graph structure into a composite risk assessment algorithm. The article concludes by describing that with the new technologies, the graph representation may become the new standard for the data analysis in the cloud environment, especially for the insurance area, stressing the pivotal role of GNNs in navigating the complexities of interconnected, dynamic data and advocating for their continued research and development to unlock even greater potentials across various sectors.

Innovations in Project Monitoring Tools for Large-Scale Infrastructure Projects

Innovations in project monitoring tools are paramount for the effective execution of large-scale infrastructure projects. This study aims to explore the transformative role of these tools in enhancing project performance, accuracy, and efficiency. The research employs a comprehensive review of existing literature, case studies, and the latest advancements in project monitoring practices. The study highlights that integrating digital tools such as Building Information Modeling (BIM), unmanned aerial vehicles (UAVs), and Geographic Information Systems (GIS) significantly improves project coordination and real-time decision-making. Automated progress monitoring technologies reduce reliance on manual data entry, enhancing the accuracy of project assessments. Additionally, sensor-based monitoring techniques enable predictive maintenance strategies, extending the lifespan of aging infrastructure. The findings underscore the importance of interoperability among digital tools and the need for robust monitoring models capable of adapting to the dynamic nature of infrastructure projects. Web-based applications and systems engineering principles further enhance the efficacy of project monitoring by providing real-time data and facilitating efficient communication among stakeholders. In conclusion, the integration of innovative project monitoring tools is pivotal for the successful execution and management of large-scale infrastructure projects. These technologies not only enhance project performance and reduce risks but also ensure sustainable and long-term infrastructure development. The study recommends continued research and investment in advanced monitoring systems to address existing challenges and unlock new opportunities for improving project outcomes. Embracing these innovations will be crucial for the construction industry to meet future demands and achieve higher standards of efficiency and sustainability. Keywords: Building Information Modeling (BIM), Unmanned Aerial Vehicles (UAVs), Geographic Information Systems (GIS), Automated Progress Monitoring, Predictive Maintenance, Project Performance Optimization.

Mitigating housing market shocks: an agent-based reinforcement learning approach with implications for real-time decision support

ABSTRACT Research in modelling housing market dynamics using agent-based models (ABMs) has grown due to the rise of accessible individual-level data. This research involves forecasting house prices, analysing urban regeneration, and the impact of economic shocks. There is a trend towards using machine learning (ML) algorithms to enhance ABM decision-making frameworks. This study investigates exogenous shocks to the UK housing market and integrates reinforcement learning (RL) to adapt housing market dynamics in an ABM. Results show agents can learn real-time trends and make decisions to manage shocks, achieving goals like adjusting the median house price without pre-determined rules. This model is transferable to other housing markets with similar complexities. The RL agent adjusts mortgage interest rates based on market conditions. Importantly, our model shows how a central bank agent learned conservative behaviours in sensitive scenarios, aligning with a 2009 study, demonstrating emergent behavioural patterns.

A dynamic two-phased approach for planning downstream oil supply chain network under uncertainty

This paper introduces an innovative approach for designing a downstream oil supply chain network under uncertain conditions, aiming to bridge existing gaps in tactical decision-making and day-to-day operational management. In the initial phase, the focus lies on tactical and operational decisions related to network planning, while the second phase emphasizes real-time transshipment decisions for daily operations. This two-phased strategy proves to be adaptable and proficient in effectively handling uncertainties. Applying the mixed possibilistic-flexible programming and a rolling horizon approach, we enhance the overall performance of the oil supply chain network. To validate the efficacy of our proposed method, we conduct a comprehensive case study utilizing a database sourced from governmental agencies, specifically tailored to the Iranian oil supply chain. Our results demonstrate the efficiency of the proposed approach in addressing uncertainties, including demand fluctuations and network costs. Furthermore, the methodology exhibits potential for broader applications in other supply chain systems and industries grappling with similar challenges.

Applications of Fog Computing in Healthcare.

Fog computing is a decentralized computing infrastructure that processes data at or near its source, reducing latency and bandwidth usage. This technology is gaining traction in healthcare due to its potential to enhance real-time data processing and decision-making capabilities in critical medical scenarios. A systematic review of existing literature on fog computing in healthcare was conducted. The review included searches in major databases such as PubMed, IEEE Xplore, Scopus, and Google Scholar. The search terms used were "fog computing in healthcare," "real-time diagnostics and fog computing," "continuous patient monitoring fog computing," "predictive analytics fog computing," "interoperability in fog computing healthcare," "scalability issues fog computing healthcare," and "security challenges fog computing healthcare." Articles published between 2010 and 2023 were considered. Inclusion criteria encompassed peer-reviewed articles, conference papers, and review articles focusing on the applications of fog computing in healthcare. Exclusion criteria were articles not available in English, those not related to healthcare applications, and those lacking empirical data. Data extraction focused on the applications of fog computing in real-time diagnostics, continuous monitoring, predictive analytics, and the identified challenges of interoperability, scalability, and security. Fog computing significantly enhances diagnostic capabilities by facilitating real-time data analysis, crucial for urgent diagnostics such as stroke detection, by processing data closer to its source. It also improves monitoring during surgeries by enabling real-time processing of vital signs and physiological parameters, thereby enhancing patient safety. In chronic disease management, continuous data collection and analysis through wearable devices allow for proactive disease management and timely adjustments to treatment plans. Additionally, fog computing supports telemedicine by enabling real-time communication between remote specialists and patients, thereby improving access to specialist care in underserved regions. Fog computing offers transformative potential in healthcare, improving diagnostic precision, patient monitoring, and personalized treatment. Addressing the challenges of interoperability, scalability, and security will be crucial for fully realizing the benefits of fog computing in healthcare, leading to a more connected and efficient healthcare environment.

Frequency security-constrained unit commitment with fast frequency support of DFIG-based wind power plants

The integration of inverter-based renewable energy sources into power grids presents a unique challenge: a reduction in grid inertia, leading to potential frequency stability issues. This paper introduces an innovative approach to address this challenge, focusing on the dynamic role of wind power plants equipped with dual-fed induction generators (DFIG). Unlike traditional methods that rely on fixed inertia control strategies and wind power droop coefficients, our method offers a dynamic solution tailored to the fluctuating nature of grid frequency dynamics. We propose a novel variable inertia support mechanism, coupled with an optimized power point tracking control and strategic reserve, to maximize the effective use of wind power reserves without waste. Additionally, we account for the inherent uncertainties in wind power generation, enhancing the frequency support capabilities of wind turbines through a fast frequency response strategy. A key feature of our approach is the application of an advanced piecewise linearization fitting method to establish a robust frequency nadir constraint. This is integrated into a two-stage robust frequency security-constrained unit commitment model, enabling optimal real-time decision making for frequency stability in wind power plants. The efficacy and practicality of our proposed method are verified through a detailed case study on a modified IEEE 39-bus system and a hardware-in-the-loop experiment. Compared with the traditional fixed parameter control method, the cost is reduced by approximately 0.42%, while the linearization error at frequency nadirs remains within the range of -4% to 1.5%.

Influence of Geopolitical Events on Procurement Strategies and Supply Chain Dynamics in Multinational Corporations

Purpose: The aim of the study was to examine the influence of geopolitical events on procurement strategies and supply chain dynamics in multinational corporations Methodology: This study adopted a desk methodology. A desk study research design is commonly known as secondary data collection. This is basically collecting data from existing resources preferably because of its low cost advantage as compared to a field research. Our current study looked into already published studies and reports as the data was easily accessed through online journals and libraries. Findings: The study found that geopolitical events exert a profound influence on procurement strategies and supply chain dynamics within multinational corporations (MNCs), compelling them to adapt swiftly to global uncertainties. The study underscored that events such as trade conflicts, sanctions, and political instability disrupt established supply chains, necessitating strategic adjustments in sourcing, supplier relationships, and operational practices. MNCs respond by diversifying sourcing locations, enhancing supply chain transparency, and integrating robust risk management frameworks to mitigate geopolitical risks effectively. Unique Contribution to Theory, Practice and Policy: Institutional Theory, Resource Dependence Theory & Transaction Cost Economics (TCE) may be used to anchor future studies on influence of geopolitical events on procurement strategies and supply chain dynamics in multinational corporations. Encourage MNCs to adopt agile procurement strategies that prioritize flexibility and responsiveness to geopolitical disruptions. This includes establishing multi-sourcing strategies, leveraging local suppliers, and implementing digital platforms for real-time risk monitoring and decision-making. Strengthen supplier relationship management practices to build resilient supply networks. Foster transparent communication and collaboration with suppliers to co-develop contingency plans and mitigate supply chain risks arising from geopolitical events. Advocate for transparent and stable trade policies that reduce uncertainty and facilitate global supply chain operations. Engage in constructive dialogues with policymakers to address trade barriers, tariffs, and regulatory inconsistencies that hinder cross-border trade and procurement efficiency.

Developing a customised set of evidence-based quality indicators for measuring workplace violence towards healthcare workers: a modified Delphi method

BackgroundWorkplace violence (WPV) is a complex global challenge in healthcare that can only be addressed through a quality improvement initiative composed of a complex intervention. However, multiple WPV-specific quality indicators...

Artificial Intelligence Applied to Drone Control: A State of the Art

The integration of Artificial Intelligence (AI) tools and techniques has provided a significant advance in drone technology. Besides the military applications, drones are being increasingly used for logistics and cargo transportation, agriculture, construction, security and surveillance, exploration, and mobile wireless communication. The synergy between drones and AI has led to notable progress in the autonomy of drones, which have become capable of completing complex missions without direct human supervision. This study of the state of the art examines the impact of AI on improving drone autonomous behavior, covering from automation to complex real-time decision making. The paper provides detailed examples of the latest developments and applications. Ethical and regulatory challenges are also considered for the future evolution of this field of research, because drones with AI have the potential to greatly change our socioeconomic landscape.

Adoption of Precision Agriculture Technologies among Farmers: A Comprehensive Review

Precision agriculture (PA) technologies have spearheaded a revolution in contemporary farming practices, ushering in an era of sustainability, efficiency, and data-driven agricultural production. This comprehensive examination aims to delve deeply into the intricate dynamics surrounding the adoption of precision agriculture technologies among farmers, synthesizing existing literature, identifying key determinants, barriers, and implications. The adoption of precision agriculture technologies is influenced by a myriad of factors spanning socio-economic, institutional, technological, psychological, and attitudinal dimensions. Socio-economic factors such as farm size, income level, access to credit, and demographic characteristics are pivotal in shaping farmers' decisions to adopt PA technologies. Institutional factors, including government policies, extension services, market structures, and support networks, also wield significant influence over adoption patterns. Technological considerations, such as the complexity, compatibility, and perceived benefits of precision agriculture technologies, impact farmers' adoption decisions. Moreover, psychological factors such as risk perception, innovativeness, and attitudes towards change contribute to shaping farmers' willingness to embrace new technologies. The adoption of precision agriculture technologies holds immense promise for enhancing farm productivity, sustainability, and profitability. By enabling precise management of inputs such as water, fertilizers, pesticides, and seeds, PA technologies help optimize resource use, minimize environmental impact, and maximize yield potential. Furthermore, the integration of advanced data analytics, remote sensing, and IoT (Internet of Things) devices facilitates real-time monitoring, decision-making, and predictive analytics, empowering farmers with actionable insights for improved farm management. However, despite the compelling benefits, the adoption of precision agriculture technologies faces several barriers and challenges. High initial investment costs, limited access to capital, inadequate infrastructure, and technical knowledge gaps pose significant hurdles for farmers, particularly those operating in resource-constrained settings. Concerns regarding data privacy, security, and ownership also contribute to farmers' apprehensions about adopting PA technologies, raising questions about data governance and regulatory frameworks. To overcome these barriers and promote the widespread adoption of precision agriculture technologies, concerted efforts are needed from various stakeholders. Farmer education and training programs play a crucial role in enhancing digital literacy, technical skills, and awareness about the potential benefits of PA technologies. Government policies and incentives, including subsidies, grants, and tax incentives, can help alleviate financial barriers and incentivize adoption. Moreover, collaboration between researchers, industry stakeholders, and policymakers is essential to drive technological innovation, develop user-friendly solutions, and address emerging challenges. Investments in infrastructure, such as rural broadband connectivity and sensor networks, are also imperative to ensure the seamless integration and scalability of precision agriculture technologies across diverse farming landscapes.

Impact of Telemedicine on Patient-Centered Outcomes in Pediatric Critical Care: A Systematic Review

Background: Pediatric intensive care units (ICUs) face shortages of intensivists, posing challenges in delivering specialized care, especially in underserved regions. While studies on telecritical care in the adult ICU have demonstrated decreased complications and mortality, research on telemedicine in the pediatric ICU setting remains limited. This systematic review evaluates the safety and efficacy of audiovisual telemedicine in pediatric ICUs, assessing patient-centered outcomes when compared to in-person intensivist care. Methods: Two reviewers independently assessed studies from PubMed, MEDLINE (Ovid), Global Health, and EMBASE on the pediatric population in the ICU setting that were provided care by intensivists via telemedicine. Studies without a comparison group of in-person intensivists were excluded. Selected studies were graded using the Newcastle–Ottawa scale and the Levels of Evidence Rating Scale for Therapeutic Studies. Results: Of the 2419 articles identified, 7 met the inclusion criteria. Strong evidence suggested that telemedicine increases access to intensive care. Moderate evidence demonstrated that telemedicine facilitates real-time clinical decision-making, reliable remote clinical assessments, improved ICU process measures (i.e., days on a ventilator, days on antibiotics), and decreased length of stay. Weaker evidence supported that telemedicine decreases complications and mortality. Conclusions: Telemedicine may serve as a promising solution to pediatric ICUs with limited intensivist coverage, particularly in low-resource rural and international settings.

Natural Language Processing (NLP) Accurately Identifies LTCF Exposure from Clinical Notes: A Proof-of-Principle Study

Background: Residence or recent stay in a long-term care facility (LTCF) is one of the most important risk factors for multidrug-resistant organism (MDRO) carriage and infection, making reliable identification of LTCF-exposed inpatients a critical priority for infection control day-to-day practice and research. However, because most hospital electronic health records (EHRs) do not include a dedicated field for documenting LTCF exposure, absent manual review of patient charts, identifying LTCF-exposed inpatients is challenging. We aimed to develop an automated, natural language processing (NLP)-based classifier for identifying LTCF exposure from clinical notes. Methods: We randomly sampled 1020 adult admissions from 2016-2021 across the 12-hospital University of Maryland Medical System and manually reviewed each admission’s history & physical (H&P) note for mention of LTCF exposure (Figure 1). After H&P pre-processing, we calculated feature representations for documents based on term frequencies and visually explored between-group (LTCF-exposed vs. LTCF-unexposed) feature differences. To predict LTCF status from the H&P notes, we trained and tuned a LASSO regression-based classifier on 70% of the data with 3-fold cross-validation and 1:1 up-sampling to address class imbalance. The final classifier was evaluated on the 30% held-out sample (not up-sampled), with calculation of the C-statistic (area-under-the-curve, AUC) with bootstrapped 95% confidence intervals, and construction of receiver-operating-characteristic and variable importance plots (R Version 4.3.2). Results: 7% (n=76 cases) of H&P notes documented LTCF exposure. In our visual analysis, the H&P words and phrases that were over-represented among LTCF patients had high face validity (Figure 2). The final LASSO-regression-based classifier achieved a C-statistic of 0.89 (95% CI: 0.80–0.98) on the held-out data for identifying LTCF exposure from the H&P notes (Figure 3). The most important model predictors (i.e., words) for distinguishing LTCF-exposed from LTCF-unexposed patients are reflected in Figure 4. The most important predictor-words of LTCF-exposure were “rehab,” “place,” “status,” “egd,” and “dementia.” Conclusion: In this multi-center study, even a simple NLP classifier demonstrated very strong discrimination for identifying LTCF exposure status from H&P notes, which could substantially reduce the manual review time required to identify LTCF-exposed inpatients. If automated in the electronic health record, it could also inform real-time MDRO screening decisions. Future research is planned to build more sophisticated classifiers using machine learning best practices, to build classifiers for additional MDRO risk factors, and to externally validate NLP classifiers on notes from an external healthcare system.

Development and application of a qualitative rapid analysis framework in a hybrid trial within primary care

ContextIn the context of iterative feedback loops to support real-time policy decision making, and an emphasis on speeding up adoption of evidence-based interventions, qualitative healthcare researchers are increasingly expected to...

A DRL-Based Reactive Scheduling Policy for Flexible Job Shops With Random Job Arrivals

In real-life production systems, arrivals of jobs are usually unpredictable, which makes it necessary to develop solid reactive scheduling policies to meet delivery requirements. Deep reinforcement learning (DRL) based scheduling methods are capable of quickly responding to dynamic events by learning from the training data. However, most of policy networks in DRL algorithms are trained to choose priority dispatching rules (PDR), thus, to some extent, the efficiency of obtained scheduling plans is limited by the performance of PDRs. This paper investigates a dynamic flexible job shop scheduling problem with random job arrivals for the total tardiness minimization. A DRL-based reactive scheduling method, proximal policy optimization with attention-based policy network (PPO-APN), is proposed to make real-time decisions for the dynamic scheduling environment, where the attention-based policy network (APN) is able to directly select pending jobs distinguished from the action space that consists of PDRs. Additionally, a global/local reward function (GLRF) is designed to address the reward sparsity issue during training processes. The proposed PPO-APN is tested on randomly generated instances with different production configurations, and is compared with frequently-used PDRs and DRL-based methods. Numerical experimental results indicate that APN and GLRF components significantly improve the training efficiency, and the PPO-APN shows better overall performance compared with other methods. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work is motivated by a typical production scenario in discrete manufacturing systems, where orders randomly arrive at the shop floor and require to be scheduled in a short time to ensure the on-time delivery. Previous research work tends to apply DRL algorithms to choose suitable dispatching rules for the ease of implementation. Nevertheless, the jobs that can be selected by dispatching rules are rather limited, thus many possible high-quality scheduling plans are ignored. This work first sorts all the unscheduled jobs by a heuristic algorithm, and puts some of top-ranked jobs to a pool. When a machine becomes available, it will directly choose a job from the pool as the next processing task. The job selection policy is represented by a novel attention-based network, and is trained by a powerful DRL algorithm. The aforementioned process is repeatedly executed in a simulation environment to collect the training data. Therefore, after being trained for a certain period of time, the policy will become smarter and can be applied to make right decisions in real-time. The proposed reactive scheduling method has been proved to be more efficient than dispatching rules and DRL-based approaches, and is effective in the production scheduling for a wide variety of discrete manufacturing scenarios, such as automobile and electronics industries. Moreover, the proposed method can be further extended to address dynamic scheduling problems with some production characteristics via adding constraints for the job selection or re-defining calculations for the completion time of operations accordingly.

Supply Chain Optimization: Machine Learning Applications in Inventory Management for E-Commerce

This study delves into the potential impact of machine learning (ML) on supply chain optimization and inventory management for e-commerce. Our primary focus is analyzing the accuracy of demand forecasting, optimizing inventory levels, and evaluating the impact of real-time decision-making on supply chain efficiency. Using a secondary data-based review methodology, this study explores the implementation of advanced predictive analytics, real-time data processing, autonomous operations, and personalized customer experiences in prominent e-commerce companies like Amazon, Walmart, and Alibaba. Our findings show impressive advancements in demand forecasting accuracy, dynamic inventory management, and operational responsiveness. These improvements have led to cost savings and increased customer satisfaction. Nevertheless, some drawbacks exist, such as the significant expenses associated with implementation, concerns about data privacy, and the potential for overfitting the model. Policy implications call for solid data protection regulations, financial assistance for smaller businesses, and ethical guidelines for AI to promote fair and responsible machine learning applications. By tackling these obstacles, companies can maximize the potential of ML technologies to enhance efficiency, promote sustainability, and gain a competitive edge in the ever-changing world of e-commerce.

MANAGEMENT ACCOUNTING SYSTEMS (MAS): A CONCEPTUAL FRAMEWORK TO ENHANCE SME’s PERFORMANCE

Small and medium-sized enterprises (SMEs) play a vital role in the global economy by creating jobs and fostering economic expansion. SMEs, however, frequently have difficulties in optimizing their operations, especially in the areas of management accounting, financial management and decision-making. The adoption and use of Management Accounting Systems (MAS) in SMEs is still lacking, despite their critical role in providing relevant data for decision-making. This conceptual paper offers an approach for developing and carrying out successful MAS that is especially appropriate for SMEs. Using theoretical ideas and prior research, the paper outlines the fundamental components of MAS with an emphasis on SMEs, including cost accounting, budgeting, performance evaluation, and strategic planning. It is stressed that SMEs have unique characteristics that necessitate a flexible and adaptable approach to MAS deployment. These characteristics include resources constraints, limited expertise, and unpredictable environments. Furthermore, the paper explores the potential benefits of implementing modern technologies in order to improve the usability and accessibility of MAS for SMEs. By utilizing these technologies, SMEs can enhance data accuracy, enable real-time decision-making, and can overcome traditional barriers of MAS adoption. In conclusion, this conceptual paper adds to the body of literature by providing a thorough framework for creating and executing MAS that are specific to the requirements and limitations of SMEs. The proposed framework seeks to promote sustainable growth, innovation, and competitiveness in the SME sector by strengthening the financial management capacities of SMEs.

Predicting tablet properties using In-Line measurements and evolutionary equation Discovery: A high shear wet granulation study

High shear wet granulation (HSWG) is widely used in tablet manufacturing mainly because of its advantages in improving flowability, powder handling, process run time, size distribution, and preventing segregation. In line process analytical technology measurements are essential in capturing detailed particle dynamics and presenting real-time data to uncover the complexity of the HSWG process and ultimately for process control.This study presents an opportunity to predict the properties of the granules and tablets through torque measurement of the granulation bowl and the force exerted on a novel force probe within the powder bed. Inline force measurements are found to be more sensitive than torque measurements to the granulation process. The characteristic force profiles present the overall fingerprint of the high shear wet granulation, in which the evolution of the granule formation can improve our understanding of the granulation process. This provides rich information relating to the properties of the granules, identification of the even distribution of the binder liquid, and potential granulation end point. Data were obtained from an experimental high shear mixer across a range of key process parameters using a face-centred surface response design of experiment (DoE). A closed-form analytical model was developed from the DOE matrix using the discovery of evolutionary equations. The model is able to provide a strong predictive indication of the expected tablet tensile strength based only on the data in-line. The use of a closed form mathematical equation carries notable advantages over other AI methodologies such as artificial neural networks, notably improved interpretability/interrogability, and minimal inference costs, thus allowing the model to be used for real-time decision making and process control. The capability of accurately predicting, in real time, the required compaction force required to achieve the desired tablet tensile strength from upstream data carries the potential to ensure compression machine settings rapidly reach and are maintained at optimal values, thus maximising efficiency and minimising waste.

Distributed resource optimisation using the Q-learning algorithm, in device-to-device communication: A reinforcement learning paradigm

In the context of wireless systems going forward, particularly in the Beyond 5G (B5G) era, where high data rates and low latency are critical, D2D communication is a pivotal technology that has many advantages. In D2D communication, the allocation of resources plays a critical role in achieving higher throughput while ensuring interference management, improved spectrum and energy efficiency, and system fairness. Conventional resource allocation methodologies encounter challenges in dynamically changing and diverse communication environments. To deal with the dynamic and unpredictable character of channel characteristics, we utilize a distributed iterative resource allocation technique based on the reinforcement learning (RL) approach that empowers the system to learn and adapt to the wireless environment autonomously. In this paper, we formulate a distributed Q learning-based RL method as its real-time learning capabilities, reduced communication overhead, and exploration efficiency make it well-suited for adapting to dynamic D2D environments compared to other RL methods. By applying the Q-learning method, the D2D devices act as learning agents striving to maximize the cumulative rewards. Through interactions with the environment and continuous learning from feedback, these agents adapt to real-time resource allocation decisions over time. Comparing our proposed Q-learning method to the state-of-the-art RL techniques, simulation results show improvements in energy and spectrum efficiency, latency, an increase in Jain's fairness index, and improvements in overall system throughput of about 6 %–8 %. The scalability is found to be 1.69 interpreting that Q-learning exhibits a good scalability as the throughput does not abruptly decrease for an increasing number of devices.

A novel edge computing approach to astronomical image data processing based on sCMOS camera using SoC

The ever-growing deluge of astronomical data challenges traditional server-based processing, hindering real-time analysis and scientific discovery. This paper proposes a novel approach: edge computing directly on an sCMOS camera using a System-on-Chip (SoC) architecture currently developed at Creotech Instruments. We present a custom-designed camera equipped with an FPGA-based SoC, enabling on-board preprocessing and feature extraction of astronomical images. This significantly reduces data transmission, minimizes latency, and empowers real-time decision-making for critical observations. We showcase the camera's capabilities through real-world scenarios, demonstrating its usability in astronomy.

Wayfinding in pairs: comparing the planning and navigation performance of dyads and individuals in a real-world environment

Navigation is essential to life, and it is cognitively complex, drawing on abilities such as prospective and situated planning, spatial memory, location recognition, and real-time decision-making. In many cases, day-to-day navigation is embedded in a social context where cognition and behavior are shaped by others, but the great majority of existing research in spatial cognition has focused on individuals. The two studies we report here contribute to our understanding of social wayfinding, assessing the performance of paired and individual navigators on a real-world wayfinding task in which they were instructed to minimize time and distance traveled. In the first study, we recruited 30 pairs of friends (familiar dyads); in the second, we recruited 30 solo participants (individuals). We compare the two studies to the results of an earlier study of 30 pairs of strangers (unfamiliar dyads). We draw out differences in performance with respect to spatial, social, and cognitive considerations. Of the three conditions, solo participants were least successful in reaching the destination accurately on their initial attempt. Friends traveled more efficiently than either strangers or individuals. Working with a partner also appeared to lend confidence to wayfinders: dyads of either familiarity type were more persistent than individuals in the navigation task, even after encountering challenges or making incorrect attempts. Route selection was additionally impacted by route complexity and unfamiliarity with the study area. Navigators explicitly used ease of remembering as a planning criterion, and the resulting differences in route complexity likely influenced success during enacted navigation.