Many pressurized water distribution systems use pumps for the transport of water and tank filling. Modelling groups of parallel pumps with a common control target remains an open problem in hydraulic modelling. In this article, the authors show how to model flow- and pressure-controlled pumping stations in the analysis of hydraulic pipe networks. The process comprises two distinct phases. In the first phase, the pump station is regarded as a single surrogate link connected to the remainder of the network. The flow and head gain at the active pump stations are computed to ensure satisfaction of the network load requirements. In the second phase, an energy minimization problem is formulated for each local pump station to ascertain the optimal pump speed and which pumps should be active. For real-time applications, very significant improvements are possible by hybrid modelling, such as coupling deterministic modelling, surrogate modelling and neural networks. This can lead to performance improvement with a magnitude of the order of $ {10}^5 $. The application to optimal pump scheduling in the context of strongly varying electricity tariffs is summarized.

Ambiguity, in the decision-theoretic sense, means that agents are unable to identify unique probabilities for some events that they care about. Ambiguity characterizes many real-life situations, but many important questions surrounding it are still open. Descriptively, we know that people typically perceive and are sensitive to ambiguity in certain kinds of situations. Intuitively, this is well justified. Normatively, however, many think that ambiguous beliefs and ambiguity sensitivity are irrational. This raises questions such as: Why are people sensitive to ambiguity? Does it lead to inferior decisions, in particular given people’s usual decision environments? An interesting clue is that there are many examples of social contexts in which ambiguity benefits everyone involved. Hence, we investigate the possibility that ambiguity sensitivity is ‘ecologically rational’ or adaptive in a multi-agent, strategic setting. We explore the viability of ambiguity sensitive behaviour using evolutionary simulations. Our results indicate that ambiguity sensitivity can be adaptive in strategic contexts, and is especially beneficial when agents have to coordinate.

Type 2 diabetes (T2D) incidence has been steadily increasing over the past few decades. Several studies have evaluated the effect of plant-based, vegetarian or vegan diets on the risk of T2D, although their potential benefits need to be confirmed and characterised. We performed a literature search up to 10 July 2025, using the terms/keywords related to plant-based index (PDI), vegetarian/vegan diets and T2D. We included observational non-experimental studies evaluating adherence to such diets in adult subjects assessing T2D risk. We specifically considered overall PDI and related healthy PDI (hPDI) and unhealthy PDI (uPDI), assessing intake of different food groups. We included 36 studies published between 1999 and 2025. We found an inverse association between adherence to vegetarian/plant-based dietary patterns and T2D risk. This association was stronger, though statistically imprecise, for the vegan diet (RR = 0·65, 95 % CI 0·42, 1·00) and for lacto-ovo-vegetarian diet (RR = 0·68, 0·57, 0·82). For studies using plant-based indices, the RR were 0·82 (0·69, 0·82), 0·76 (0·69, 0·82) and 1·13 (0·98, 1·30) for overall PDI, hPDI and uPDI, respectively. In the dose–response meta-analysis, overall PDI and hPDI showed an inverse and almost linear association with T2D risk. Conversely, adherence to uPDI directly correlated with T2D risk. Overall, adherence to vegan/vegetarian diets may reduce T2D risk, while an unhealthy plant-based diet appears to linearly increase disease risk, indicating caution in the consumption of such unhealthy foods even if of plant origin. The beneficial association between vegetarian and healthy plant-based diets may have major public health implications.

The rise of large language models (LLMs) has marked a substantial leap toward artificial general intelligence. However, the utilization of LLMs in (re)insurance sector remains a challenging problem because of the gap between general capabilities and domain-specific requirements. Two prevalent methods for domain specialization of LLMs involve prompt engineering and fine-tuning. In this study, we aim to evaluate the efficacy of LLMs, enhanced with prompt engineering and fine-tuning techniques, on quantitative reasoning tasks within the (re)insurance domain. It is found that (1) compared to prompt engineering, fine-tuning with task-specific calculation dataset provides a remarkable leap in performance, even exceeding the performance of larger pre-trained LLMs; (2) when acquired task-specific calculation data are limited, supplementing LLMs with domain-specific knowledge dataset is an effective alternative; and (3) enhanced reasoning capabilities should be the primary focus for LLMs when tackling quantitative tasks, surpassing mere computational skills. Moreover, the fine-tuned models demonstrate a consistent aptitude for common-sense reasoning and factual knowledge, as evidenced by their performance on public benchmarks. Overall, this study demonstrates the potential of LLMs to be utilized as powerful tools to serve as AI assistants and solve quantitative reasoning tasks in (re)insurance sector.

Prenatal maternal mental health and social determinants of health may influence pregnancy, child hospitalisation, and child neurodevelopmental outcomes in critical congential heart disease (CHD). We examined 189 mother–child dyads of children born with CHD who underwent neonatal cardiac surgery and completed neurodevelopmental assessment between the ages 13 and 29 months. We used latent profile analysis to identify distinct maternal groups based on prenatal maternal mental health screening scores and individual- and neighbourhood-level social determinants of health factors. We examined the association between maternal groups with their child’s gestational age, birth weight, hospital length of stay (HLOS), and neurodevelopment. Latent profile analysis identified two distinct groups: high-risk (n = 46) and low-risk (n = 143). Mothers in the high-risk group had higher mental health screening scores, lower age, higher social vulnerability, lower education, and were more likely to have Medicaid insurance and represent a minority group than mothers in the low-risk group. The high-risk group had children with lower gestational age and weight at birth, longer HLOS, and lower cognitive, language, and motor scales than children in the low-risk group (p < 0.05). Sensitivity analysis in mother–infant dyads without foetal extracardiac conditions found that significant relationships persisted in the high-risk group, with lower gestational age and lower language scale scores than the low-risk group (p < 0.05). Children of mothers with adverse prenatal maternal mental health and social determinants of health risks had significantly worse pregnancy and child outcomes. Interventions are critically needed to address maternal mental health and social determinants of health risks beginning in the prenatal period.

A five-year-old boy was diagnosed intraoperatively with the Raghib complex, consisting of an absent coronary sinus, persistent left superior vena cava, and an atrial septal defect associated with a ventricular septal defect. The surgical management, which included intracardiac rerouting of the left superior vena cava to the right atrium along with closure of both septal defects, was successfully performed.

The presence of glufosinate-resistant Palmer amaranth (Amaranthus palmeri S. Watson) is of concern for Arkansas farmers. The objective of this study was to understand the distribution of glufosinate resistance among A. palmeri accessions collected in 2023 from locations surrounding MSR2 (a highly glufosinate-resistant accession) in 2020, focusing on the distance and direction patterns. Additionally, the cytosolic (GS1) and chloroplastic (GS2) glutamine synthetase copy number were quantified in glufosinate survivors. In 2023, a total of 66 A. palmeri samples were collected within a 15-km radius of MSR2. Amaranthus palmeri seedlings were treated with glufosinate at 590 g ai ha−1. Plant tissues were collected, and gene copy number assays were conducted with survivors from accessions showing less than 96% mortality. Glufosinate provided ≥80% mortality in most of the accessions evaluated. Nonetheless, a few accessions showed low mortality rates, with values as low as 34%. Within and among accessions, there was no variation for GS1.1 and GS1.2, while the GS2.1 and GS2.2 copy numbers varied greatly. There was no evidence that the geographic distance between samples and MSR2 impacted mortality or gene copy number. However, there was strong evidence that direction, relative to MSR2, affected both mortality and GS2.1 copies. Samples collected north from MSR2 showed lower average mortality rates (83%) with a higher number of GS2.1 copies (2.3). For comparison, average mortality ranged from 90% to 95% and GS2.1 copy number ranged from 1 to 1.2 in the other directions. The predominant summer and fall wind directions do not explain the movement of resistance in a specific direction. These findings indicate that there are multiple A. palmeri accessions capable of surviving a label recommended use rate of glufosinate in northeast Arkansas, and resistance distribution needs to be further investigated.

In two field experiments conducted in Mississippi and Florida, we present novel evidence about how emotions can be harnessed to increase voter turnout. When we inform respondents that a partisan villain would be happy if they did not vote (for example, a Gloating Villain treatment), we find that anger is activated in comparison to other emotions and turnout increases by 1.7 percentage points. In a subsequent field experiment, we benchmark this treatment to a standard GOTV message, the social pressure treatment. Using survey experiments that replicate our field experimental treatments, we show that our treatment links the act of voting to anticipated anger. In doing so, we contribute the first in-the-field evidence of how we can induce emotions, which are commonly understood to be fleeting states, to shape temporally distant political behaviours such as voting.

Strengthening the research workforce is essential for meeting the evolving needs and challenges in the health and biomedical fields. To do so effectively, it requires an understanding of how the experiences of a researcher shift over time and how one’s research career evolves, particularly as supports are put in place to foster research. This narrative review provides a summary of published individual-level assessment measures and survey tools from 2000–2024. All measures were abstracted, classified, and coded during analyses to describe the areas of focus, and they were organized into one of six research categories. The review identified a range of measures and methods across all categories. However, the measures were often narrow, focused on outputs, and not ideal for assessing the full range of experiences a researcher may have throughout their career. The most common metrics were related to research productivity and bibliometric measures. Our review of survey tools revealed a gap in comprehensive approaches available to assess an individual’s research experience, efforts, supports, and impact. As efforts expand to evaluate and study the research workforce, tools that focus on a broad range of individual-level measures, tied to specific underlying constructs and drawn from the literature, may prove useful.

We consider local escape rates and hitting time statistics for unimodal interval maps of Misiurewicz–Thurston type. We prove that for any point z in the interval, there is a local escape rate and hitting time statistics that is one of three types. While it is key that we cover all points z, the particular interest here is when z is periodic and in the postcritical orbit that yields the third part of the trichotomy. We also prove generalized asymptotic escape rates of the form first shown by Bruin, Demers and Todd.

The adoption of upper limb myoelectric prosthesis is limited by the lack of closed control loop systems. Although the efferent control has already been integrated into these devices, the sensory feedback restoration in the afferent channel still remains an open challenge. Transcutaneous electrical nerve stimulation (TENS) is a promising method for generating somatotopic sensory feedback, allowing the closure of the control loop system. The application of this technique is limited by cumbersome and grid-powered electrical stimulators, making them unsuitable for everyday life, whereas most portable stimulators available on the market are designed for other purposes (e.g., muscular stimulation or pain therapy) and present limited stimulation wave customization. The stimulation devices employed in the literature often produce not fully suitable stimulation parameters and are frequently validated through procedures that do not fully clarify their practical application for sensory feedback restoration. The research aims to present a novel wearable TENS stimulation device (46 g, 62 × 49 × 20 mm) suitable for sensory feedback application. The validation was achieved through a benchtop test and a preliminary analysis on 10 healthy participants comparing the qualities, intensities, and stimulated areas of the sensations elicited by the proposed device and a reference stimulator. The proposed device is capable of delivering charge-balanced stimulation waves over skin-like resistive load and eliciting tingling and vibration sensations with similar intensities compared to the adopted reference.

Bilinguals with aphasia routinely experience anomia in one or both of their languages that may be ameliorated by language treatment. Traditionally, treatment response has been captured by binary scoring systems that measure the presence or absence of improvement without examining how word retrieval attempts may change over time as a function of treatment. This study analyzed word retrieval errors and subsequent treatment outcomes for a group of 48 Spanish-English bilinguals with aphasia to determine if longitudinal error patterns could capture language recovery. Results revealed naming improvement for trained words in the treated language and translations of trained words in the untreated language. Specific types of word errors at baseline were associated with overall improvement in both languages; furthermore, patterns of responses changed over time as a function of lexical-semantic treatment. These results demonstrate that error analyses may characterize bilingual treatment outcomes and provide new evidence for mechanisms of impaired word retrieval.