Segmented Model Research Articles

Study on the smoke flow characteristics in inclined tunnel with an upstream shaft under natural ventilation

The rapid expansion of urban highway tunnels has led to a significant increase in the risk of fire incidents. Moreover, the urban highway tunnels often have inclinations, which further exacerbates the complexity of the smoke flow. In contrast to the extensive research on natural ventilation in horizontally tunnels, there is still a lack of studies on the natural ventilation of shaft in inclined tunnels. Therefore, this study employed numerical simulations and theoretical analysis to investigate the length of smoke back-layering and flow patterns in inclined tunnel fires with an upstream shaft. By comparing the induced air inflow velocity and smoke back-layering length in inclined tunnels, both with and without an upstream shaft, three types of smoke flow states were identified. State 1: smoke spreading to the upstream shaft, State 2: smoke being limited between the shaft and the fire source, State 3: no smoke backflow in the inclined tunnel. In State 1, the shaft discharges smoke outward. The downstream length and the distance between the shaft and the fire source determined smoke back-layering length. In States 2 and 3, the shaft fills the air inward. The shaft has no obvious effect on the smoke back-layering length. Furthermore, the effects of HRR, fire source position and shaft position on smoke flow characteristics are analyzed. As a result of the analysis, a segmented prediction model for smoke back-layering length and a quantitative method for predicting smoke flow state are proposed. By fine-tuning the tunnel slope, two critical values between the three smoke flow states are obtained. The research results can provide a useful reference for ventilation and evacuation design of inclined tunnels.

Association Between Prescription Drug Monitoring Program Use Mandate and Opioid Prescribing and Patient-Reported Outcomes After Surgery.

To evaluate changes in opioid prescribing and patient-reported outcomes after surgery following implementation of Michigan's prescription drug monitoring program (PDMP) use mandate in June 2018. Most states mandate clinicians to query prescription drug monitoring program (PDMP) databases before prescribing controlled substances. Whether these PDMP use mandates affect opioid prescribing and patient-reported outcomes after surgery is unclear, especially among patients with elevated "Narx" scores, a risk score for overdose death used in most PDMPs. We conducted an interrupted time series analysis of a statewide surgical registry linked to Michigan's PDMP database. Analyses included adults undergoing general surgical procedures during January 2017-October 2019. Outcomes included monthly mean days supplied in dispensed opioid prescriptions (those filled within 3 days of discharge) and monthly mean scores for 3 patient-reported outcomes (pain in the week after surgery, care satisfaction, regret undergoing surgery). Segmented regression models were used to assess for level and slope changes in outcomes in June 2018. Analyses were repeated among patients with Narx scores ≥200, a threshold that defines the top quartile. Analyses included 21,897 patients. The mandate was associated with a -0.5 (95% CI: -0.8, -0.2) level decrease in mean days supplied in dispensed opioid prescriptions, but not with worsened patient-reported outcomes. Findings were similar among patients with Narx scores ≥200. Following implementation of Michigan's PDMP use mandate, the duration of opioid prescriptions decreased, but patient-reported outcomes did not worsen. Findings suggest PDMP use mandates may not be associated with worsened experience among general surgical patients.

The impact of the English national health inequalities strategy on inequalities in mortality at age 65: a time-trend analysis.

During the 1997-2010 Labour government, several policies were implemented to narrow health inequalities as part of a national health inequalities strategy. Many of these policies are likely to have had a disproportionately large impact on people aged 65 and over. We aimed to understand the association between the health inequalities strategy period and inequalities in mortality at age 65-69. We use population at risk and mortality data covering 1991-2019 to calculate mortality rate at age 65-69 at the Local Authority level. We use the 2019 Index of Multiple Deprivation to examine geographical inequalities. We employ segmented linear regression models with marginal spline terms for the strategy period and interact these with an indicator of deprivation to understand how inequalities changed before, during and after the strategy. The reporting of this study adheres to STROBE guidelines. Mortality rates in each deprivation quintile improved continuously throughout the period of study. Prior to the programme (1991-9) there was no significant change in absolute inequalities. However, during the strategy (2000-10) there was a significant decrease in absolute inequalities of -9.66 (-17.48 to -1.84). The period following the strategy (2011-19) was associated with a significant increase in absolute inequalities of 12.84 (6.60 to 19.08). Our results were robust to a range of sensitivity tests. The English health inequalities strategy was associated with a significant reduction in absolute inequality in mortality age 65-69. Future strategies to address inequalities in ageing populations may benefit from adopting a similar approach.

Abstract PO5-16-06: Prophylactic Neurokinin-1 Receptor Antagonist Use Pre- and Post- Choosing Wisely Initiative among Women with Breast Cancer

Abstract Objective: In April 2012, the American Society of Clinical Oncology (ASCO) issued recommendations as a part of the Choosing Wisely (CW) campaign with the aim to identify the areas of low-value utilization. One of the CW recommendations is not to use expensive antiemetics, specifically Neurokinin-1 Receptor Antagonist (NK1-RA) for patients initiating low or moderate emetic risk chemotherapy. The purpose of this study was to determine the impact of CW on prophylactic NK1-RA use among women with breast cancer. Methods: Using Optum’s de-identified Clinformatics® Data Mart Database for 2010 to 2018, a retrospective cohort study was conducted for women aged ≥18 years with breast cancer with at least one newly initiated claim for low/minimal/moderate emetic risk chemotherapy (n = 25,549). All approved antiemetics - aprepitant, fosaprepitant, rolapitant, and netupitant/palonosetron) were included. An interrupted time series using a segmented regression model was used to assess the effect of CW on the prophylactic NK1-RA use. Results: The prophylactic use of NK1-RAs among patients with breast cancer receiving low/minimal/moderate emetic risk chemotherapy decreased from 11.1% pre-CW to 7.7% post-CW. Segmented regression analysis showed that there was a significant increase of 0.11 per 100 patients per quarter in the use of prophylactic NK1-RAs prior to CW recommendation. However, immediately after the CW (occurred in Q42013), there was a significant decline in prophylactic NK1-RA use by 1.03 per 100 patients in Q22014. For the time after intervention, there was a significant decline in NK1-RA use by 0.37 per 100 patients per quarter in the post-CW period compared to the pre-CW period (p< 0.05). Conclusion: This study highlights a significant but modest decline in the inappropriate use of prophylactic NK1-RAs after the implementation of CW antiemetic recommendation. Continued statewide or local educational efforts for dissemination of CW recommendations are needed to facilitate appropriate prophylactic use of NK1-RAs. Table. Adjusted* Interrupted Time Series Segmented Regression Analysis of Prophylactic NK1-RA use among women with breast cancer before and after the Choosing Wisely initiative *Adjusted for insurance, health plan type, age, comorbidity, and emetic risk Citation Format: Shweta Kamat, Britny Brown, Steven Cohen, Ami Vyas. Prophylactic Neurokinin-1 Receptor Antagonist Use Pre- and Post- Choosing Wisely Initiative among Women with Breast Cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-16-06.

Suicide mortality following the implementation of tobacco packaging and pricing policies in Korea: an interrupted time-series analysis

BackgroundTo prevent tobacco use in Korea, the national quitline number was added to tobacco packages in December 2012, tobacco prices were raised by 80% in January 2015, and graphic health warning labels were placed on tobacco packages in December 2016. This study evaluated the association of these tobacco packaging and pricing policies with suicide mortality in Korea.MethodsMonthly mortality from suicide was obtained from Cause-of-Death Statistics in Korea from December 2007 to December 2019. Interrupted time-series analysis was performed using segmented Poisson regression models. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated adjusted for suicide prevention strategies.ResultsSuicide mortality was 20 per 1,000,000 in December 2007 and showed a downward trend over the study period. After the implementation of tobacco packaging and pricing policies, suicide mortality immediately declined by − 0.09 percent points (95% CI = − 0.19 to 0.01; P > 0.05) for the national quitline number, − 0.22 percent points (95% CI = − 0.35 to − 0.09; P < 0.01) for tobacco prices, and − 0.30 percent points (95% CI = − 0.49 to − 0.11; P < 0.01) for graphic health warning labels. The corresponding RRs for these post-implementation changes compared with the pre-implementation level were 0.91 (95% CI = 0.83 to 1.00), 0.80 (95% CI = 0.70 to 0.91), and 0.74 (95% CI = 0.61 to 0.90), respectively. Significant associations between tobacco control policies and suicide mortality were observed even when stratified by sex and region.ConclusionsThe findings of this study provide new evidence for an association between tobacco control policies and deaths by suicide. An array of effective tobacco control policies should be considered for prevention programs targeting suicide.

Pirarucu requires taurine to maximize growth and antioxidant status when fed diets high in plant-based ingredients

Taurine (Tau) is a free conditionally essential amino acid for most fish species that plays a significant nutritional role, mainly in carnivorous fish species. However, limited studies are still available regarding the role of Tau on pirarucu nutrition, an important Amazonian fish species. Therefore, this study used diets high in plant-based ingredients to investigate the effect of taurine (Tau) supplementation on juvenile pirarucu's growth, metabolism, and health (n = 115, Weight = 504.12 ± 19 g). Fish were fed five diets containing increasing levels of Tau: 0.95, 5.1, 9.6, 12.9, and 17.2 g/kg in a completely randomized design with three replicates per treatment. After a 60-day feeding trial, growth, hematology, blood chemistry, Tau content in muscle, carcass, and liver, antioxidant status, and immunological parameters were evaluated. The control group (0.95 g/kg Tau) showed the lowest growth. According to the segmented regression model, the recommended Tau level for maximizing the development of juvenile pirarucu was 4.49 g/kg of feed. Additionally, fish fed the lowest dietary Tau level showed the highest ROS (reactive oxygen species) concentration. Dietary Tau levels did not affect hematological parameters and blood chemistry (p > 0.05). Fish fed 9.6 g/kg of Tau showed increased catalase activity, while the control group had the worst result in NBT (nitroblue tetrazolium reduction test). Fish fed with the highest dietary Tau levels (12.9 and 17.2 g/kg) showed increased total serum immunoglobulin concentration. These results suggest that the optimal inclusion level of Tau in diets for juvenile pirarucu is 4.49 g/kg of feed and that this amino acid acts as a mitigator of oxidative stress.

Effects of design and fabrication of segmented ship models on test results of wave-induced loads

The design and fabrication of segmented ship models have great effects on wave-induced loads in wave basins. Three different segmented ship models of a 20,000 TEU (twenty-feet equivalent unit) ultra-large container ship are designed and fabricated with different support frames and backbones. The shell and wooden frames of Models A and B were fabricated with the same method, however, the type of backbone and scale are different. Model C is a steel framing model, the dimensions of the frame members are validated by finite element method and the frame members are connected with welding technique. Suitable parameters are selected to design and fabricate Model C with the improved fabrication method. The calculated mode results of Model C are compared with the experimental data to validate the reasonability and advantage of the improved design and fabrication method. The dry and wet modes of the three ship models are measured and compared. The three different ship models are tested under typical conditions in different wave basins, and the corresponding test results are compared to reveal the effects of the fabrication method on structural elastic modes and wave-induced loads. Finally, the uncertainty analysis of the three model tests under typical conditions are conducted.

The study of novelty nutrients supplementary scheme and methodology on advanced carbon sequestration of microalgae

Microalgae has great potential in carbon sequestration and production of biofuel even with a low concentrations of CO2. In this paper, an Andrews based microalgae growth kinetic model and a novel and multiple nutrient supplementary scheme that significantly extends the carbon sequestration period and enhances growth rates of microalgae are proposed. First, the correlation between nutrient concentration and microalgal growth is constructed experimentally. The advanced growth kinetic model incorporating nitrogen and phosphorus consumption is developed. Then, the optimum cultivation conditions including initial nutrients concentration, nutrients supplementary interval and daily nutrients supplementary quantities are obtained in the conical flask reactor. The findings led to a novel nutrient supplementation strategy that initiates with a 0.2BC concentration, followed by a 0.3BC addition at 100 h, with subsequent supplements every 72 h. After adopting this supplementary strategy, the microalgae growth rate is increased by 42.9 % and the carbon sequestration rate is increased by 56.2 % compared with that of the original microalgae cultivation method in one cultivation cycle. Finally, the universal microalgae nutrients supplementary model is proposed in a piecewise model. It can be concluded from the model that in the first cultivation period, microalgae needs a higher affinity for nutrients and after nutrients supplementary, the sensitivity to nutrient inhibition is reduced. The paper is significantly helpful to improve the sustainability of the microalgae growth rate and the carbon sequestration rate.

Prediction of Delayed Surface Subsidence Based on the Improved Knothe-n Model

The delayed surface subsidence caused by coal seam mining is a problem that cannot be ignored, while accurate prediction of the surface subsidence provides a guarantee of the safety and stability of the relevant areas. However, the traditional Knothe model has limitations in considering delayed surface subsidence. Because of this, the Knothe-n time function model is segmented and improved by using the data of the subsidence area obtained from a FLAC3D-based numerical model, and the maximum delayed surface subsidence in different periods is calculated. The analytical results are compared with the numerical results to validate the effectiveness of the improved segmented time function model in predicting delayed surface subsidence. The improved model is applied to predict the surface subsidence in the Yutianbao subsidence area. The root-mean-square error between the predicted and measured values for the maximum subsidence monitoring point is 1.12, and the root-mean-square error between the average predicted and measured values for the surface monitoring points is 0.37, which verifies the accuracy of the improved model. The prediction model provides a scientific basis for environmental protection and safety management after coal seam mining.

ОЦІНКА СТРУКТУРИ ШЛАКОВИХ РОЗПЛАВІВ НА ПІДСТАВІ АНАЛІЗУ ТЕМПЕРАТУРНИХ ЗАЛЕЖНОСТЕЙ В’ЯЗКОСТІ ТА ЕЛЕКТРОПРОВІДНОСТІ З ВИКОРИСТАННЯМ АДАПТИВНОЇ СЕГМЕНТНОЇ РЕГРЕСІЙНОЇ МОДЕЛІ

The expediency of using an adaptive segmented regression model for the analysis of logarithmic dependences of viscosity and electrical conductivity on temperature to assess the structure of slag melts is shown. Calculated values of the activation energies of viscosity ( ) and electrical conductivity ( ) in different temperature ranges. The obtained calculation results confirm that the viscosity and electrical conductivity of slag melts characterize the interaction of various structural particles (ions and/or their groups) and can be used in the study of the structure of slag melts. It was established that the activation energy of viscosity and electrical conductivity decreases as the temperature of slag melts increases. The performed calculations of the indicator n=En/Ex indicate that it can be a criterion for evaluating the structure of the slag melt, when n &gt;1, the slag melt is a heterogeneous system, and when n &lt; 1, it is a homogeneous system. Viscosity and electrical conductivity are interdependent properties and can be predicted if there is data on one of the properties, provided the value of n is known.

Experimental and numerical investigation on the pyrolysis of oil shale particles in a bubbling fluidized bed

The efficient utilization of oil shale resource is beneficial to environment and the resource diversity. In this work, a bubbling fluidized bed (BFB) reactor was first used to pyrolyze Huadian oil shale experimentally, mainly focusing on the influence of fluidization flow rates. The yield of shale oil, char and non-condensable gas varies little at different fluidization flow rates, but the components in shale oil are different due to the secondary cracking, such as the fractions and species of aliphatic hydrocarbons, aromatic hydrocarbons and non-hydrocarbons. Then, the dense discrete phase model (DDPM) was used to track the trajectories of oil shale particles with different diameters in a bubbling fluidized bed. Population balance model (PBM) coupled Eulerian model was used to well describe the bubbling status of bed materials to characterize the hydrodynamic behaviors. Furthermore, a segmented kinetic model was compiled by UDF module to illuminate the pyrolysis of oil shale particles. The multi-phase reactive flow modeling clearly indicated the distribution of oil shale particles and gaseous products in BFB reactor, as well as pyrolysis time and the secondary cracking at different fluidization flow rates. In summary, this work provides a reference basis for complete pyrolysis of oil shale particles in a bubbling fluidized bed and generating shale oil with high yield and good quality.

Experimental analysis of bow flare slamming and whipping responses in a ship at different sailing speeds

This study conducted an experimental investigation using a segmented model to determine bow flare slamming loads and whipping responses of a ship in regular waves. The model simulated the distribution of the vertical and horizontal bending stiffnesses of the hull. A continuous wavelet transform was employed to capture temporal variations in the harmonic components. The influences of sailing speeds on the temporal and spatial variations in the slamming pressures in the bow flare were analyzed. Subsequently, the time-frequency characteristics and asymmetry of the global responses under different test conditions are discussed. The random errors in the experimental system were investigated through repeated testing. The findings revealed that a higher sailing speed could significantly influence the magnitude and distribution of slamming pressures on the bow flare, while playing an important role in inducing asymmetric slamming loads under oblique waves. The slamming-induced whipping responses under regular waves caused considerable periodic changes in the strengths of the harmonic components around the wet natural frequency of the model. At high sailing speeds, the whipping responses significantly increased the asymmetry of the vertical bending moment (VBM) at midship.

Changes in the Appropriateness of US Outpatient Antibiotic Prescribing After the COVID-19 Outbreak: An Interrupted Time Series Analysis of 2016-2021 Data.

No national study has evaluated changes in the appropriateness of US outpatient antibiotic prescribing across all conditions and age groups after the coronavirus disease 2019 (COVID-19) outbreak in March 2020. This was an interrupted time series analysis of Optum's de-identified Clinformatics Data Mart Database, a national commercial and Medicare Advantage claims database. Analyses included prescriptions for antibiotics dispensed to children and adults enrolled during each month during 2017-2021. For each prescription, we applied our previously developed antibiotic appropriateness classification scheme to International Classification of Diseases, Tenth Revision, Clinical Modification diagnosis codes on medical claims occurring on or during the 3 days prior to dispensing. Outcomes included the monthly proportion of antibiotic prescriptions that were inappropriate and the monthly proportion of enrollees with ≥1 inappropriate prescription. Using segmented regression models, we assessed for level and slope changes in outcomes in March 2020. Analyses included 37 566 581 enrollees, of whom 19 154 059 (51.0%) were female. The proportion of enrollees with ≥1 inappropriate prescription decreased in March 2020 (level decrease: -0.80 percentage points [95% confidence interval {CI}, -1.09% to -.51%]) and subsequently increased (slope increase: 0.02 percentage points per month [95% CI, .01%-.03%]), partly because overall antibiotic dispensing rebounded and partly because the proportion of antibiotic prescriptions that were inappropriate increased (slope increase: 0.11 percentage points per month [95% CI, .04%-.18%]). In December 2021, the proportion of enrollees with ≥1 inappropriate prescription equaled the corresponding proportion in December 2019. Despite an initial decline, the proportion of enrollees exposed to inappropriate antibiotics returned to baseline levels by December 2021. Findings underscore the continued importance of outpatient antibiotic stewardship initiatives.

Development of New Predictive Equations for the Resting Metabolic Rate (RMR) of Women with Lipedema.

This study aimed to develop a novel predictive equation for calculating resting metabolic rate (RMR) in women with lipedema. We recruited 119 women diagnosed with lipedema from the Angiology Outpatient Clinic at Wroclaw Medical University, Poland. RMR was assessed using indirect calorimetry, while body composition and anthropometric measurements were conducted using standardized protocols. Due to multicollinearity among predictors, classical multiple regression was deemed inadequate for developing the new equation. Therefore, we employed machine learning techniques, utilizing principal component analysis (PCA) for dimensionality reduction and predictor selection. Regression models, including support vector regression (SVR), random forest regression (RFR), and k-nearest neighbor (kNN) were evaluated in Python's scikit-learn framework, with hyperparameter tuning via GridSearchCV. Model performance was assessed through mean absolute percentage error (MAPE) and cross-validation, complemented by Bland-Altman plots for method comparison. A novel equation incorporating body composition parameters was developed, addressing a gap in accurate RMR prediction methods. By incorporating measurements of body circumference and body composition parameters alongside traditional predictors, the model's accuracy was improved. The segmented regression model outperformed others, achieving an MAPE of 10.78%. The proposed predictive equation for RMR offers a practical tool for personalized treatment planning in patients with lipedema.

Exploration of anatomical distribution of brain metastasis from breast cancer at first diagnosis assisted by artificial intelligence

ObjectivesThis study aimed to explore the spatial distribution of brain metastases (BMs) from breast cancer (BC) and to identify the high-risk sub-structures in BMs that are involved at first diagnosis. MethodsMagnetic resonance imaging (MRI) scans were retrospectively reviewed at our centre. The brain was divided into eight regions according to its anatomy and function, and the volume of each region was calculated. The identification and volume calculation of metastatic brain lesions were accomplished using an automatically segmented 3D BUC-Net model. The observed and expected rates of BMs were compared using 2-tailed proportional hypothesis testing. ResultsA total of 250 patients with BC who presented with 1694 BMs were retrospectively identified. The overall observed incidences of the substructures were as follows: cerebellum, 42.1 %; frontal lobe, 20.1 %; occipital lobe, 9.7 %; temporal lobe, 8.0 %; parietal lobe, 13.1 %; thalamus, 4.7 %; brainstem, 0.9 %; and hippocampus, 1.3 %. Compared with the expected rate based on the volume of different brain regions, the cerebellum, occipital lobe, and thalamus were identified as higher risk regions for BMs (P value ≤ 5.6*10−3). Sub-group analysis according to the type of BC indicated that patients with triple-negative BC had a high risk of involvement of the hippocampus and brainstem. ConclusionsAmong patients with BC, the cerebellum, occipital lobe and thalamus were identified as higher-risk regions than expected for BMs. The brainstem and hippocampus were high-risk areas of the BMs in triple negative breast cancer. However, further validation of this conclusion requires a larger sample size.

Recognition of concrete microcrack images under fluorescent excitation based on attention mechanism deep recurrent neural networks

This study aims to recognize fluorescent excitation images of microcracks in concrete components through the adoption of an Attention Mechanism-based Deep Recurrent Neural Network (RNN) model, thereby enhancing the accuracy and efficiency of crack detection. Considering the significance of concrete crack detection and the limitations in efficiency and accuracy of existing methods, this paper proposes an innovative image processing technique that combines fluorescent excitation methods with deep learning models to achieve earlier and more accurate identification of concrete microcracks. C30 concrete specimens were prepared experimentally and treated with fluorescent solution spray. Fluorescent images of cracks under UV light excitation were collected and processed using a segmented attention mechanism deep RNN model. Various performance evaluation metrics, including mean Intersection over Union (mIoU), mean Image Intersection over Union (miIoU), mean Image Dice Coefficient (miDice), and F1 score, were employed to comprehensively assess the model's performance. The results demonstrate that the proposed model achieved significant effectiveness in concrete crack image recognition, showing higher mIoU, miIoU, miDice, and F1 scores compared to other representative deep learning models, thus proving its advantages in recognition accuracy and efficiency. Particularly, by introducing the segmented attention mechanism, the model could capture microcrack features more effectively, significantly improving the accuracy of crack identification. This method not only provides a new technical approach for the early detection of concrete cracks but also lays the foundation for further development of efficient and accurate crack detection technologies to accommodate more complex engineering application scenarios.

Rapid Estimation Model for Wake Disturbances in Offshore Floating Wind Turbines

The precise wake model is crucial for accurately estimating wind farm loads and power, playing a key role in wake control within wind farms. This study proposes a segmented dual-Gaussian wake model, which is built upon existing dual-Gaussian wake models but places greater emphasis on the influence of initial wake generation and evolution processes on the wind speed profile in the near-wake region. The enhanced model optimizes the wake speed profile in the near-wake region and improves the accuracy of wake diffusion throughout the entire flow field. Furthermore, the optimized dual-Gaussian wake model is utilized to estimate the power output and blade root vibration loads in offshore wind farms. Through comparative analysis of high-fidelity simulation results and actual measurement data, the accuracy of the optimized dual-Gaussian wake model is validated. This approach offers high computational efficiency and provides valuable insights for load fluctuations and power estimation, thereby advancing the development of wake control strategies rapidly.

Hybrid black-box classification for customer churn prediction with segmented interpretability analysis

Customer retention management relies on advanced analytics for decision making. Decision makers in this area require methods that are capable of accurately predicting which customers are likely to churn and that allow to discover drivers of customer churn. As a result, customer churn prediction models are frequently evaluated based on both their predictive performance and their capacity to extract meaningful insights from the models. In this paper, we extend hybrid segmented models for customer churn prediction by incorporating powerful models that can capture non-linearities. To ensure the interpretability of such segmented hybrid models, we introduce a novel model-agnostic approach that extends SHAP. We extensively benchmark the proposed methods on 14 customer churn datasets on their predictive performance. The interpretability aspect of the new model-agnostic approach for interpreting hybrid segmented models is illustrated using a case study. Our contributions to decision making literature are threefold. First, we introduce new hybrid segmented models as powerful tools for decision makers to boost predictive performance. Second, we provide insights in the relative predictive performance by an extensive benchmarking study that compares the new hybrid segmented methods with their base models and existing hybrid models. Third, we propose a model-agnostic tool for segmented hybrid models that provide decision makers with a tool to gain insights for any hybrid segmented model and illustrate it on a case study. Although we focus on customer retention management in this study, this paper is also relevant for decision makers that rely on predictive modeling for other tasks.

Regional differences in stem form between southern and northern red spruce (Picea rubens Sarg.) populations

Abstract Red spruce (Picea rubens Sarg.) is a commercially and ecologically important conifer species that primarily exists at northern latitudes of eastern North America and extends southward following the Appalachian Mountains into North Carolina and Tennessee. Due to a warming climate and human-caused disturbances, only fragmented, sky-island populations remain at the highest peaks of the southern Appalachians where their habitat continues to be threatened. While they have been recognized for the rare wildlife habitat they provide in the region, these populations remain understudied. This work aimed to examine differences in stem form between the northern and southern populations of red spruce and to provide additional quantitative methods for managing red spruce stands through providing regionally fitted stem taper models. First, we examined differences in stem form using two methods: a sectional rate of change in diameter and a region variable added to the Kozak (2004) Model 02 taper model. The sectional taper comparison showed significant differences (P &amp;lt; .05) in taper rate throughout the stem that were most pronounced below breast height and above the midpoint. The nested model comparison also showed a significant difference after performing a likelihood ratio test. These results agreed that significant differences in stem form between the two populations exist and supported the idea that localized taper models would provide the best results. Next, we evaluated four stem taper models for their ability to predict upper stem diameters and total volume in southern Appalachian red spruce: a quadratic polynomial, a segmented, a variable exponent, and a geometric model. Fitting data came from a 1992 study across three southern Appalachian spruce-fir sites. Validation data came from a new dataset collected in 2022 at Unaka Mountain, Tennessee. Based on past studies and our results, we found that the Kozak (My last words on taper equations. For Chron 2004;80:507–15. https://doi.org/10.5558/tfc80507-4.) Model 02 variable exponent and the Max and Burkhart (Segmented polynomial regression applied to taper equations. For Sci 1976;22:283–9.) segmented polynomial models performed equally well. The choice of the final model should depend on the users’ objectives and practical limitations (i.e. programming ability, availability of fitting data, error tolerance).

Study on stress–strain curve and damage evolution model of expanded perlite concrete under uniaxial compression

AbstractThis study investigates the influence of expanded perlite (EP) replacement rate and curing methods on the compressive strength of concrete at different ages through the uniaxial compression test. The stress–strain full curve model of concrete considering EP replacement rate is established, and a damage evolution model considering curing conditions, EP replacement rate and load effects is developed based on the damage theory. The results show that: (1) The lack of water in concrete curing will have a huge negative impact on its mechanical properties, and the internal curing effect of pre‐wet EP can effectively alleviate the phenomenon of incomplete hydration. (2) Under dry curing, the addition of EP will reduce the early compressive strength of concrete, while its internal curing effect can improve the compressive strength of concrete in the later stages. (3) The C &amp; C model and the Guozhenhai model are used to construct the segmented model in the rising and falling sections, and the model parameter formula is deduced by the replacement rate. The model can fully describe the stress–strain relationship of expanded perlite concrete (EPC). (4) The damage evolution model of EPC shows an “S” shape growth curve. Dry curing and EP replacement can both reduce the development rate of damage, while the former has a more significant impact.