ABSTRACT Current Earth-observing task planning of satellites highly depends on ground decision-making and control which is implemented in a two-stage manner, i.e. ground planning then satellite implementation. While on-board satellite computing capabilities have made autonomous in-orbit planning more feasible, this is still focused on solving traditional planning problems with limited tasks or handling emergent tasks with on-board re-planning, limiting further situation-based autonomy required by large constellation management and control. While situation-aware in-orbit planning is unaddressed, this paper proposed a novel situation representation and fast planning algorithm to tackle batch tasks planning and re-planning together in a single algorithm framework. This is implemented by first representing the tasks as heatmaps which is quantified and aligned using grid geographical transformations. This situation representation is then handled to guide the decision of satellite sequential footprint movements as an accurate allocation of satellite resources. The autonomy is enhanced by modelling the situation dynamic adaption problem as a high-dimensional Markov Dynamic Process which is then solved efficiently using deep reinforcement learning. The experiments have demonstrated the efficacy of the proposed algorithm in the provision of high-quality planned solution and high efficiency.

Child welfare notifications are often interpreted as indicators of need, yet they originate at the system's "front door," where actors differ in roles, access, and reporting conventions. We therefore examine notifications as intake signals: who reports what, and when. To map linkages between reporter groups and stated reasons, identify temporal rhythms in notifications, and test short-run coupling between reporting channels. All child-welfare notifications filed in one Finnish wellbeing services county between 1 Oct 2023 and 30 Sep 2024 (N=5715). Notifications were coded by reporter group and reason category. Reporter-reason patterns were analyzed using cross-tabulations, adjusted standardized residuals, and correspondence analysis. Daily and weekly time series were constructed to assess weekday-weekend and school-holiday rhythms. Short-run lead-lag relations between reporter groups were tested using Granger screening and HAC-robust lagged OLS models. Findings reflect system-level predictive dynamics rather than causal effects or case-level sequences. Reporter groups exhibited a clear child-to-service gradient. Emergency and safety authorities predominantly cited child-related reasons, whereas anonymous contacts mainly concerned service-related issues. Schools displayed a mixed profile, while welfare professionals and relatives more often emphasized family-related concerns. Notification volumes followed pronounced weekday and school-calendar rhythms, with declines during holidays. Weekly models showed inverse short-run coupling: increases in school reports preceded decreases in emergency-service reports, and similar patterns appeared across other reporter pairs. Notifications reflect an institutionally structured reporting ecology rather than a uniform indicator of need. Divergent framing and inverse timing across channels suggest that volume changes can signal shifts in visibility and access at intake, with implications for interpretation, resource allocation, and prevention planning.

Fairness during resource allocation influences event-related potential (ERP) responses during memory of receivers' faces.

Optimization of maintenance planning in power distribution systems using a discrete-time Markov chain model asset analysis and resource allocation

This study examines the COVID-19 experiences of K-12 school campus directors, especially in terms of building and grounds, transportation, and food services from the Midwestern United States. The COVID-19 pandemic has changed to online education mode and implemented safety protocols to ensure equitable access to K-12 education. School directors play a crucial role in United States’s school administration. The researchers used a mixed method research design and purposive sampling. The sample included 9 school-grounds directors, 34 food service directors, and 31 transportation directors. A 5-point Likert-type questionnaire was prepared by the researchers to collect the relevant data for the study. The data were analyzed using descriptive statistics and thematic analysis. While revealing their experiences, various school directors mentioned that key challenges are reliable sources of information for them to work, inequities in resource allocation, finance allocation, and lack of professional training. The survey highlighted the need for several resources, including federal and state funding, teacher professional development programs, and community partnerships that support K-12 education. School directors addressed challenges regarding transportation, buildings, grounds, and food services during COVID-19. The study revealed the need for systemic policy changes and communication resources to create resilient and adaptive learning environments for K-12 education. Eventually, the study recommended that policymakers and educational leaders bring changes to K-12 infrastructure to withstand crises like COVID-19.

Strawberry ( Fragaria × ananassa ) production in controlled environments often exhibits cyclical fruiting patterns, or “flushes,” which complicate labor, resource, and market planning. This study aimed to develop a yield forecasting tool for greenhouse strawberry production using flower mapping, a technique that characterizes floral bud development through meristem dissection. Using a soilless culture system with hanging gutters, we grew a widely used cultivar Albion in a greenhouse maintained at 22.5 ± 3.1 °C daytime and 18.2 ± 2.8 °C nighttime air temperatures, 20.0 ± 3.0 mol·m −2 ·d −1 daily light integral, 16-hour photoperiod, 580 ± 207 µmol·mol −1 daytime CO 2 , and 1.0 ± 0.6 kPa. Plants were grown in a commercial strawberry substrate composed of 100% coconut coir fiber and irrigated with a nutrient solution delivered by a drip irrigation system. Plants were transplanted on 17–18 Aug 2023 and flower mapping occurred weekly on randomly sampled plants during a 19-week crop cycle from 21 Jan to 26 May 2024. We found the number of primary buds at floral developmental stage 3 (when secondary buds differentiate) exhibited a significant positive correlation with weekly fruit yield occurring 11 weeks later. In addition, we found the number of primary buds at floral developmental stages 4 and 5 (when calyx and trichomes differentiate on the floral bud, respectively) exhibited significant positive correlations with weekly yield occurring 9 weeks later. In addition, stage 11 flower buds (flowers at anthesis) showed a significant positive correlation with yield occurring 3 weeks later. The remaining developmental stages exhibited weaker correlations and were less reliable predictors of upcoming yield. In addition, a similar correlation analyses performed using fruit number per plant as the outcome variable showed consistent results with the yield-based analyses but with generally weaker correlations, indicating that fruit number may be more variable than yield. Key developmental stages 3, 4, 5, and 11 can be used for developing a methodology for forecasting near-future yield. This will help US greenhouse strawberry growers to make informed decisions about resource allocation, labor scheduling, and market planning, ultimately optimizing yield and production efficiency in controlled environment agriculture systems.

Fractional-order safe mental-health corridor modelling with Matignon spectral analysis of post-pandemic fatigue-to-recovery dynamics.

Nursing workload in pediatric intensive care units is complex and increasingly demanding. Effective staffing is essential for positive patient outcomes, as inadequate coverage correlates with higher mortality and readmission rates. Current staffing tools have limitations and fail to account for the unique challenges of pediatric care. To develop and validate a workload prediction tool for pediatric intensive care nurses to enhance decision-making and resource allocation. The QuantI2S tool was developed through literature review and expert consensus, then implemented in a 24-bed Pediatric Intensive Care Unit. Validation involved: 1) correlation with the current gold standard, 2) inter-rater reproducibility, and 3) predictive accuracy. The bedside nurse and clinical nurse specialist completed the QuantI2S two hours before shift end (prospective score), while an independent reviewer calculated a retrospective score from chart reviews. Agreement was assessed using Bland-Altman plots and Intra-Class Correlation (ICC). A total of 172 patient-observations involving 45 patients were analyzed (July-August 2016). Compared with the gold standard, QuantI2S showed excellent reliability (rs=0.738, 95% CI [0.624-0.822], p=0.001). ICC for prospective scores was strong (0.990, 95% CI [0.985-0.993]). Bland-Altman analysis revealed near-perfect agreement (mean difference-0.03). Prospective and retrospective scores also showed excellent concordance (ICC=0.916, 95% CI [0.872-0.946]). QuantI2S accurately predicts pediatric intensive care nursing workload, demonstrating excellent reliability and ease of use. By integrating nursing activities and child-specific factors, it provides a robust framework for optimizing staffing and improving patient care.

Prompt-to-policy: Leveraging large language models to guide deep reinforcement learning in public health emergencies.

Machine learning-based immunocompromise and severity score for early risk stratification of critically ill patients.

National and subnational burden of XDR-TB in China,1990-2023: Long-term trends, regional disparities, and projections, 2024-2050.

This article examines the leadership strategies used by school principals to manage inclusive education for students with special educational needs. Based on a systematic review of 25 articles from top academic journals (2021-2025), the research highlights the key role principals play in promoting inclusivity in schools. Findings show that effective leadership practices, such as transformational leadership, community involvement, and resource allocation, significantly enhance the success of inclusive education. Principals who foster a culture of support and high expectations for teachers contribute to better educational outcomes for special educational needs students. The study also emphasizes the importance of collaboration between teachers, parents, and external organizations in supporting inclusive practices. However, the research identifies challenges that principals face, such as limited resources and managing diverse learning needs. Professional development and ongoing training for educators are crucial for overcoming these barriers. The article concludes that while character-driven leadership centered on integrity, empathy, and ethical decision-making plays a positive role in building an inclusive school environment, it must be balanced with practical strategies to address operational challenges. Effective leadership is essential for creating a sustainable, inclusive educational environment that meets the diverse needs of all students.

Chronic hepatitis B (CHB) refers to a global infectious disease caused by the hepatitis B virus. The treatment of CHB causes a heavy economic burden to society. To ensure the rational allocation of medical resources in the whole society and achieve the goal of patients' satisfaction and economy, this study aimed to evaluate the economics of de novo combination of tenofovir alafenamide fumarate (TAF) as the first-line nucleos(t)ide analogues (NAs) and peginterferon alfa-2b (PEG-IFNα-2b) versus PEG-IFNα-2b monotherapy of HBeAg-positive CHB in China. The Markov model was used to simulate the transition of HBeAg-positive CHB patients aged 30 in China under various health states using TreeAge Pro 2011 software. The cycle length was 1 year, and the cycle period of the model was 50 years. The model parameters included clinical efficacy, cost, transition probability and discount rate. Cost-effectiveness analysis was conducted through simulation of the total cost and quality adjusted life years (QALYs) of various treatment options through models. Simultaneously, one-way sensitivity analysis, probabilistic sensitivity analysis and scenario analysis were performed. De novo combination of TAF and PEG-IFNα-2b and PEG-IFNα-2b monotherapy resulted in 11.16 and 10.81 QALYs, with total costs of $55559.72 and $57670.23, respectively. De novo combination strategy for HBeAg-positive CHB patients can save costs and obtain more health outcomes. Sensitivity analyses showed the reliability of the results. From the perspective of the whole society, the de novo combination strategy of TAF and PEG-IFNα-2b for patients with HBeAg-positive CHB may be more cost-effective than PEG-IFNα-2b monotherapy.

Existing shipping environmental regulations largely omit the economic dimension which, in turn, delays the industry's clean energy transition. This paper investigates the efficiency of global shipping on economic foundations. We apply a stochastic frontier analysis to assess the interactions between capital, operation, earnings, and transport work, both across all major shipping segments and at an individual-vessel level. The empirical results indicate that carbon efficiency of vessel types decreases with speed. Larger vessel types produce more carbon emissions for a given level of TC earnings and costs. At an individual vessel level, higher production efficiency is observed in vessels that are newer, spend more time at sea, have installed more energy saving technologies (ESTs), and belong to companies with stronger EST investment policy. Technical and operational inefficiencies raise the total cost of owning and operating a vessel by 7%, with market price dynamics and inefficient allocation of economic resources increasing it by 25%. An increase in fuel price of 38% or a reduction in speed of 13.5% does not severely affect a vessel's overall efficiency and total cost. Policy interventions need to be carefully designed in order not to negatively impact the overall efficiency of global shipping.

Cultivated land abandonment is an important issue for the protection and management of cultivated land. It is related to farmers' livelihood endowments and the optimal allocation of cultivated land resources. Furthermore, it directly affects food security, agricultural modernization, and comprehensive rural revitalization. In this study, we identified 30 influencing factors across four categories: natural geographic factors and farming conditions, socioeconomics, farmers' overall competency, and government policies, and employed the DEMATEL-ISM-MICMAC model to illustrate the hierarchical structure, cause-and-result diagram, and driver-dependency matrix of the factors. We found that (1) the 30 influencing factors interact with each other, generating a five-level hierarchical structure, which can be classified into neighboring, transitional, and essential causes. (2) The system of factors influencing cultivated land abandonment includes 11 causal factors, 19 effect factors, and 13 critical factors. Agricultural Socialized Services have a strong driving effect on the other factors, and Distance to the Main Road, Township and County and Development Level of Land Transfer Market can influence the remaining factors. (3) These factors can be classified into four categories in terms of their driving and dependence power: autonomous, dependent, linkage, and driving factors, among which driving and linkage factors are more effective in inhibiting the occurrence of cultivated land abandonment. This study identified critical factors affecting the occurrence of cultivated land abandonment, and provides theoretical guidance for the management of cultivated land abandonment, thereby serving as practical reference for developing and improving relevant cultivated-land preservation systems and agricultural policies in the future. • Construct the influencing factor system of cultivated land abandonment • DEMATEL-ISM-MICMAC model was used to analyze the influencing factors of cultivated land abandonment. • Identify the attributes, status and importance of each influencing factor in the influencing factor system of cultivated land abandonment.

• CFO co-option is positively associated with the intensity of R&D investments. • CEO risk-taking incentives (Vega) partially mediate this relationship. • These results are more pronounced in the early years of co-option. This paper examines the influence of chief financial officer (CFO) co-option, defined as the appointment of a CFO by the incumbent chief executive officer (CEO), on research and development (R&D) investments. Drawing on the upper echelon (UE) theory, we argue that co-opted CFOs are more likely to internalize the strategic preferences for innovation of their appointing CEOs, reducing their monitoring role and amplifying the influence of CEOs over financial resource allocation, thereby facilitating greater R&D investments. We further argue that CEO risk-taking incentives serve as an important mechanism through which CFO co-option influences R&D investments because co-opted CFOs can reinforce CEO risk preferences by supporting compensation structures that incentivize risk-taking. Using a sample of 10,370 firm-year observations from 2011 to 2022, we find that CFO co-option is positively associated with the intensity of R&D investments and that this relationship is partially mediated by CEO risk-taking incentives, consistent with our arguments. These results extend the co-option literature by revealing that R&D investments are not only directly influenced through shared preferences but also indirectly through risk-taking mechanisms. Overall, these findings underscore the need for careful board oversight to balance innovation and risk.

Cloud-edge collaborative task offloading and resource allocation based on mobile computility

Development and validation of interpretable machine learning models for predicting long-term functional outcomes in elderly patients with aneurysmal subarachnoid hemorrhage.

In the management of adult moyamoya disease (MMD), there is clinical equipoise regarding direct bypass (DB) vs indirect bypass (IB). One factor that may influence the choice of procedure is cost. Therefore, we sought to compare the true intraoperative costs (not charges or reimbursement) of DB vs. IB using a novel costing methodology, time-driven activity-based costing (TDABC). A retrospective cohort study (2017-22) was conducted of all patients undergoing revascularization for symptomatic MMD. TDABC was used to calculate intraoperative costs. Supply cost was calculated as the aggregate of expenses related to implants, consumables, medications, and surgical tray sterilization. Personnel cost was determined by multiplying the per-minute wages of all intraoperative personnel by the amount of time they spent intraoperatively. Software was developed to automate this data from the electronic medical record. Multivariable regression was performed to compare the costs of DB vs. IB, adjusting for age, smoking status, hypertension, and coronary artery disease. Of 38 patients undergoing revascularization for MMD, 22 (57.9%) underwent DB, while 16 (42.1%) underwent IB. The median total cost of DB surgery was $7,802 (IQR: $7,802 - $9,881), compared to $6,120 (IQR: $5,609 - $6,900)for IB surgery (p<0.05). On descriptive analysis, DB also had higher median supply cost ($3,793 vs. $2,855, p<0.01), which was driven by higher cost of consumables ($2,268 vs. $1,996, p<0.01), surgical tray sterilization ($137 vs. $129, p<0.01), and implants ($1,022 vs. $568, p<0.05). On multivariable analysis, DB was associated with significantly higher total cost (β:$1,758±$768, p=0.027) and supply cost (β:$1,032±$333, p=0.004), with no significant differences in personnel cost (β:$726±$526,p=0.175). Using granular costing methodology, we demonstrate the individual drivers of cost which are responsible for differences in expenditures between DB and IB. This framework may inform surgical decision-making, guide institutional budgeting, and optimize resource allocation for the treatment of MMD.

In multi-intersection traffic signal timing (TST) the use of a uniform cycle time reduces efficiency at low-volume intersections and signal changes which can alter vehicle flows and degrade communication quality. This paper presents a solution to this problem introducing a Twin-Timescale model whereby a multi-intersection TST method is jointly considered with communication resource allocation (CRA) strategies. For each intersection, a Twin-Timescale Reinforcement Learning (RL) algorithm is proposed, by considering both CRA and TST as the Short- and the Long- Timescale models, respectively, and jointly formulating their rewards. Based on the characteristics of this Twin-Timescale model, the reward function is constructed as a singularly perturbed model (SPM), and a state feedback controller that can ensure asymptotic stability is designed. Furthermore, a novel asynchronous time Multi-Agent Coordinated RL algorithm is proposed, which allows each intersection to generate each own joint parameters for its decision strategy which are based upon information obtained from neighboring intersections at asynchronous time. In addition, the use of Federated Learning (FL) in conjunction with the proposed algorithms is investigated, and its advantages in enhancing the resistance to malicious attacks are discussed. Using real traffic data, various performance evaluation results obtained by means of extensive computer simulation experiments have confirmed that the proposed TST-CRA joint optimization approach significantly improves traffic indicators and maintains the quality of the overall communication performance. Through numerical calculations, the state feedback controller coefficients have been determined.