Optimal Interval Research Articles

Optimal Service Strategies of Online Platform Based on Purchase Behavior

In the rapidly evolving platform economy, online platforms have emerged as pivotal providers of digital services to sellers. The paper investigates how online platforms optimize service strategies based on consumers’ purchase behavior, influencing sellers’ pricing and social welfare. Using a two-period Hotelling model and a cooperative game framework, we discover that the optimal service strategies of a platform with data collecting capabilities are collaborating with two sellers to offer to extend services to new consumers in the second period, maximizing profits for all sellers and platform. Applying Shapley value analysis, we determine the platform’s equitable service charge strategies. When sellers adopt behavior-based pricing (BBP), the pricing escalates in the first period, and the platform’s optimal service strategies also enhance the pricing of sellers. However, in the second period, BBP intensifies competition, leading to generally lower pricing. Our findings suggest that optimal pricing in the second period for new consumers should increase with enhanced quality perception, which is provided by the platform’s digital services and heightened by consumers’ privacy concerns, while decreasing for regular consumers. Lastly, we offer policy recommendations, exploring optimal regulatory scenarios—limiting or not limiting data collection—to maximize social welfare or consumer surplus, and the Mathematica software is used to identify distinct optimal policy intervals.

Prediction of Biochar Adsorption of Uranium in Wastewater and Inversion of Key Influencing Parameters Based on Ensemble Learning.

With the rapid development of industrialization, the problem of heavy metal wastewater treatment has become increasingly serious, posing a serious threat to the environment and human health. Biochar shows great potential for application in the field of wastewater treatment; however, biochars prepared from different biomass sources and experimental conditions have different physicochemical properties, resulting in differences in their adsorption capacity for uranium, which limits their wide application in wastewater treatment. Therefore, there is an urgent need to deeply explore and optimize the key parameter settings of biochar to significantly improve its adsorption capacity. This paper combines the nonlinear mapping capability of SCN and the ensemble learning advantage of the Adaboost algorithm based on existing experimental data on wastewater treatment. The accuracy of the model is evaluated by metrics such as coefficient of determination (R2) and error rate. It was found that the Adaboost-SCN model showed significant advantages in terms of prediction accuracy, precision, model stability and generalization ability compared to the SCN model alone. In order to further improve the performance of the model, this paper combined Adaboost-SCN with maximum information coefficient (MIC), random forest (RF) and energy valley optimizer (EVO) feature selection methods to construct three models, namely, MIC-Adaboost-SCN, RF-Adaboost-SCN and EVO-Adaboost-SCN. The results show that the prediction model with added feature selection is significantly better than the Adaboost-SCN model without feature selection in each evaluation index, and EVO has the most significant effect on feature selection. Finally, the correlation between biochar adsorption properties and production parameters was discussed through the inversion study of key parameters, and optimal parameter intervals were proposed to improve the adsorption properties. Providing strong support for the wide application of biochar in the field of wastewater treatment helps to solve the urgent environmental problem of heavy metal wastewater treatment.

Digital technology and Chinese-style industrial modernization: Dynamic threshold effect based on R&D Human resources

Based on the construction of digital technology evaluation index system, this paper builds a dynamic threshold regression model to explore the complex impact of digital technology on Chinese-style industrial modernization under the threshold of R&D human resources, taking 30 provinces as research objects. It has been found that China's digital technology index is continuously improving, but there is a digital gap among regions. The beneficial impact of digital technology on Chinese-style industrial modernization has been thoroughly validated. Considering the threshold effect of R&D human resources, the influence of digital technology on Chinese-style industrial modernization exhibits nonlinear characteristics. With R&D human resources crossing the first threshold, it has shown a significant positive effect on Chinese-style industrial modernization, and the middle range of R&D human resources presents the optimal interval of the relationship between the two. The research findings offer a theoretical framework for advancing the construction of Chinese-style modernization.

Optimizing feeding frequencies in fish: A meta-analysis and machine learning approach

Feeding is a critical factor in aquaculture, directly impacting cultivation success and profitability. Feeding frequency determines feed intake and regulates fish appetite and nutrient utilization. However, the effects of increased feeding frequency on fish growth and feed conversion ratio (FCR) can vary depending on species, cultivation conditions, feeding rates. This variability can lead to inconsistencies in interpretations, with individual studies often lacking in statistical power. To address this issue, we analyzed data from 145 study cases across 53 publications to quantify the impact of feeding frequency on average daily weight gain (ADG) and FCR. Our meta-analysis revealed that increasing feeding frequency could reduce FCR, but it did not enhance ADG when daily feed amounts were fixed. Only when each feeding reached satiation did increased feeding frequency boost ADG. We found that the relationship between feeding frequency and ADG in fish is quadratic, with most fish species achieving optimal ADG at four feedings per day. Traditional approaches suggest that fish appetite recovery is linked to gastric evacuation time, defining an optimal feeding interval at 80 % gastric evacuation. However, our research indicates that this approach is not suitable for all fish species. Therefore, we developed a Gradient Boosting Machine (GBM) model that predicts FCR (R2 = 0.9303) and ADG (R2 = 0.7249) based on environmental variables, fish characteristics, culture period, and feeding frequency. This study provides a scientific basis for formulating more efficient and sustainable aquaculture feeding strategies, optimizing growth and feed utilization across different fish species.

Practical and Optimal Likelihood Intervals and Regions for Weibull and Gumbel Distributions

Profile likelihood intervals with good coverage probabilities and convenient simple expressions are presented to estimate separately and efficiently each of the parameters and quantiles of Weibull and Gumbel distributions, widely used in reliability and environmental applications. For the first time, likelihood-confidence intervals for these distributions, that were previously unavailable, can now be easily and quickly calculated with the provided R code for sample sizes n ≥ 10 . For simultaneous estimation of both parameters, likelihood-confidence regions are also provided. The proposal takes advantage of the transformation relationship between the Weibull and Gumbel distributions, of the fact that this latter distribution belongs to the Location and Scale family of distributions, and of likelihood symmetrizing parametrizations, all leading to simpler inferences about the parameters. The plot of the relative profile likelihood of a quantile of interest and its corresponding profile likelihood intervals provide decision makers with useful information in many practical applications, particularly when comparing two treatments. The proposed intervals for the Weibull shape parameter facilitate testing if an exponential distribution is reasonable for the data. The proposed methods are illustrated with practical examples.

Optimal surveillance interval for patients with branch duct intraductal papillary mucinous neoplasms of the pancreas (IPMN)

Optimal surveillance interval for patients with branch duct intraductal papillary mucinous neoplasms of the pancreas (IPMN)

Effect of vaccine dose intervals: Considering immunity levels, vaccine efficacy, and strain variants for disease control strategy

In this study, we present an immuno-epidemic model to understand mitigation options during an epidemic break. The model incorporates comorbidity and multiple-vaccine doses through a system of coupled integro-differential equations to analyze the epidemic rate and intensity from a knowledge of the basic reproduction number and time-distributed rate functions. Our modeling results show that the interval between vaccine doses is a key control parameter that can be tuned to significantly influence disease spread. We show that multiple doses induce a hysteresis effect in immunity levels that offers a better mitigation alternative compared to frequent vaccination which is less cost-effective while being more intrusive. Optimal dosing intervals, emphasizing the cost-effectiveness of each vaccination effort, and determined by various factors such as the level of immunity and efficacy of vaccines against different strains, appear to be crucial in disease management. The model is sufficiently generic that can be extended to accommodate specific disease forms.

High beam quality, high brightness dense spectral beam combining of 650 nm laser diode module

Dense spectral beam combining (DSBC) of a laser diode module at a wavelength of 650 nm was realized for what we believe to be the first time. In this paper, the optimal spectral locking interval range for the 650 nm wavelength band was obtained by stimulation. A DSBC device containing 19 single emitters was designed and built according to the optimal spectral interval. The output power of the combining beam reached 17.74 W, and the stable spectral width was 4.75 nm. The beam quality factors (M2) in the fast-axis and slow-axis directions were basically the same as that of the single emitter, and even slightly optimized in the slow-axis direction. The beam brightness of the DSBC beam was about 123.395 MW·cm−2·sr−1, which was about fourteen times higher than that of the single emitter output.

Impact of antenatal corticosteroids-to-delivery interval on very preterm neonatal outcomes: a retrospective study in two tertiary centers in Japan

BackgroundAntenatal corticosteroids (ACS) are administered to prevent neonatal complications and death in women at risk of imminent preterm birth. Internationally, the optimal interval from ACS to delivery (ACS-to-delivery interval) is within seven days; however, evidence in Asian populations specifically is limited. This study aimed to investigate the association between ACS-to-delivery interval and the incidence of neonatal complications in Japan.MethodsThis retrospective observational study enrolled singleton neonates born preterm at < 32 weeks of gestational age between 2012 and 2020 at two tertiary centers. A total of 625 neonates were divided into the following four groups according to the timing of ACS (measured in days): no ACS (n = 145), partial ACS (n = 85), ACS 1–7 (n = 307), and ACS ≥ 8 (n = 88). The following outcomes were compared between the groups: treated respiratory distress syndrome (RDS), severe intraventricular hemorrhage (IVH), treated patent ductus arteriosus (PDA), necrotizing enterocolitis, sepsis, bronchopulmonary dysplasia (BPD), treated retinopathy of prematurity (ROP), periventricular leukomalacia, and death discharge.ResultsThe ACS 1–7 group had significantly decreased adjusted odds ratios (ORs) for treated RDS (0.37 [95% confidence interval: 0.23, 0.57]), severe IVH (0.21 [0.07, 0.63]), treated PDA (0.47 [0.29, 0.75]), and treated ROP (0.50 [0.25, 0.99]) compared with the no ACS group. The ACS ≥ 8 group also showed significantly reduced adjusted ORs for RDS (0.37 [0.20, 0.66]) and treated PDA (0.48 [0.25, 0.91]) compared with the no ACS group. However, the adjusted ORs for BPD significantly increased in both the ACS 1–7 (1.86 [1.06, 3.28]) and ACS ≥ 8 groups (2.94 [1.43, 6.05]) compared to the no ACS group.ConclusionsAn ACS-to-delivery interval of 1–7 days achieved the lowest incidence of several complications in preterm neonates born at < 32 weeks of gestational age. Some of the favorable effects of ACS seem to continue even beyond ≥ 8 days from administration. In contrast, ACS might be associated with an increased incidence of BPD, which was most likely to be prominent in neonates delivered ≥ 8 days after receiving ACS. Based on these findings, the duration of the effect of ACS on neonatal complications should be studied further.

Optimal risk-assessment scheduling for primary prevention of cardiovascular disease

Abstract In this work, we introduce a personalized and age-specific net benefit function, composed of benefits and costs, to recommend optimal timing of risk assessments for cardiovascular disease (CVD) prevention. We extend the 2-stage landmarking model to estimate patient-specific CVD risk profiles, adjusting for time-varying covariates. We apply our model to data from the Clinical Practice Research Datalink, comprising primary care electronic health records from the UK. We find that people at lower risk could be recommended an optimal risk-assessment interval of 5 years or more. Time-varying risk factors are required to discriminate between more frequent schedules for high-risk people.

Phase 1 trial of intraperitoneal paclitaxel in patients with gastric adenocarcinoma and carcinomatosis or positive cytology.

The purpose of this phase 1 trial was to evaluate the safety and toxicity of repeated normothermic intraperitoneal paclitaxel (PTX) for patients with gastric cancer metastatic to the peritoneum. A Bayesian optimal interval design was used to prospectively identify the safety and tolerability of escalating doses of intraperitoneal paclitaxel at weekly treatments for 3 weeks, followed by a 1-week break, and then three additional treatments. The primary objective was to define the maximum tolerated dose. Secondary end points included safety, tolerability, and antitumor activity. A total of 25 patients were treated between January 2020 and April 2023. Five dose-limiting toxicities were observed at 100 mg/m2. Treatment-related grade 3-4 toxicity included leukopenia (32%) and neutropenia (32%). Seven patients required a schedule change to every other week treatments. The maximum tolerated dose for intraperitoneal PTX was 100 mg/m2. The peritoneum post-intraperitoneal PTX demonstrated progression in five (20%), stable disease in five (20%), improvement in 10 (40%), and not evaluable in five (20%). Eight patients (32%) had resolution of their peritoneal disease and seven (28%) underwent attempted resection. The median overall survival (OS) from the diagnosis of metastatic disease was 18.8 months and from the date of treatment initiation was 10.8 months. One-, 2-, and 3-year OS rates from the diagnosis of metastatic disease were 84%, 38%, and 25%, respectively. Paclitaxel may be safely used at intraperitoneal doses of 100 mg/m2. Neutropenia associated with weekly treatments was common. Peritoneal complete clinical response rates with multimodality therapy including PTX were promising.

Optimal Surveillance of Metachronous Gastric Lesion after Endoscopic Resection of Early Gastric Cancer.

Endoscopic resection (ER)-a minimal invasive procedure, compared to surgical gastrectomy, with the advantage of preserving the entire stomach and maintaining the patient's quality of life-is a widely used curative treatment for early gastric cancers (EGCs). Despite its advantages, such as the preservation of the whole stomach, a large area of the gastric mucosa with histologic changes such as atrophy and intestinal metaplasia remains after ER, and so does the risk of metachronous gastric cancers (MGCs). Therefore, regular surveillance endoscopy after curative ER of EGCs is important so that MGCs are detected early and so minimally invasive ER remains a treatment option. To date, the optimal interval for surveillance endoscopy after curative ER of EGCs has not been established. Therefore, this review summarizes the results of the published studies on this topic with the aim of establishing the optimal surveillance interval for early identification of MGCs. Based on my review, the median timing of MGC occurrence is within 3 years, and reports suggest biannual endoscopy during the first 3 years; however, the evidence suggests that individual patient characteristics may influence the risk of MGCs. Therefore, stratified endoscopic strategies for surveillance based on patient characteristics, such as age, family history of gastric cancer, synchronous gastric lesions, and corpus intestinal metaplasia, should be applied.

Abstract A004: A Phase 1B/2 trial of second line immunotherapy with pepinemab and avelumab for patients with metastatic pancreatic adenocarcinoma

Abstract Background: Pancreatic adenocarcinoma (PDAC) is a leading cause of cancer-related mortality with poor prognosis despite maximal therapy and is recalcitrant to approved immunotherapies. In PDAC, Semaphorin 4D (SEMA4D) expression is prognostic in patients who have undergone surgical resection and correlates with tumor infiltration of activated CD8+ T-cells. Furthermore, preclinical studies suggest that monoclonal blockade of SEMA4D supports CD8+T cells tumoral infiltration and activation, sensitizing PDAC to immune checkpoint blockade (ICB). Thus, we hypothesized that a humanized anti-SEMA4D antibody (pepinemab) may improve disease control and responsiveness to ICB in patients with PDAC. Methods: We designed a phase Ib/II single-arm clinical trial to evaluate the safety, tolerability, and efficacy of pepinemab with anti-PD-L1 (avelumab) in patients with chemotherapy-refractory PDAC (NCT05102721). The study follows a dose de-escalation schema starting at a dose combination of 20mg/kg pepinemab and 800mg avelumab every two weeks. Patient accrual for phase 1b utilizes the Bayesian Optimal Interval design targeting a dose-limiting toxicity rate of 30% or less. After 16 subjects receive a given combination dose, a Simon’s two stage assessment of futility will be undertaken with expansion to phase 2 if 2 or more subjects demonstrate response. Treatment response is assessed via RECIST 1.1 criteria with surveillance CT prior to enrollment and after completion of two cycles (8 weeks). Baseline and on-treatment tumor biopsies (after completion of one cycle, 4 weeks) are performed to analyze changes in immune, stromal, and genomic profiles to elucidate mechanisms of treatment response and failure. Patient reported outcomes are incorporated (FACT-Hep and FAACT subdomains) to assess disease-specific symptoms. Results: A total of 14 patients have been enrolled to date. Three subjects withdrew from the study prior to completing the evaluable period (four weeks of treatment) and nine subjects terminated treatment due to disease progression. Two subjects continue to receive combination treatment pending evaluation of tumor response. Baseline and on-treatment biopsies have been obtained and stored for correlative analysis from 8 evaluable subjects. One treatment-related adverse event occurred though the patient was able to complete treatment without further events. No treatment-related toxicities have been encountered. Conclusions: Combining pepinemab with avelumab appears to be safe and tolerable to date. Nine of eleven patients demonstrated disease progression, though 1 of these 9 demonstrated mixed response with some lesions showing radiologic partial and even complete response. Overall treatment response, survival data, and correlative science forthcoming. Citation Format: Luis I Ruffolo, Matthew Byrne, Bailey Hilty, Brian A Belt, Yatee Dave, Paul Burchard, Terrence Fisher, Elizabeth E. Evans, Crystal Mallow, Megan Boise, Darren Carpizo, Jen Jen Yeh, David C Linehan, Daniel Mulkerin. A Phase 1B/2 trial of second line immunotherapy with pepinemab and avelumab for patients with metastatic pancreatic adenocarcinoma [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr A004.

Parcel-Locker-Sharing Model for E-Commerce Logistics Service Providers

In the e-commerce last-mile delivery process, the asset operators (logistics service providers who own parcel locker facilities) support their delivery service with parcel lockers, while the non-asset operators (logistics service providers without parcel lockers) perform door-to-door delivery. Due to demand fluctuation, asset operators’ parcel-locker slots may be left vacant, while non-asset operators are stuck with the high-cost door-to-door service. The exclusiveness of parcel-locker usage reduces resource utilization and service efficiency in last-mile delivery. Therefore, this paper proposes a parcel-locker-sharing model in which these two parties share the parcel-locker capacity in last-mile delivery. The asset operator rents the unused parcel lockers to the non-asset operator by charging a rental fee,while the non-asset operator rents the parcel lockers for delivery to save logistics costs. The motivation of this alliance is to increase the profits of both parties and that of the total supply chain. This study establishes the supply-chain profit model for the parcel-locker-sharing framework and finds that the profit or loss depends on the comparison of the operation cost savings and delivery-cost savings. A numerical analysis is conducted to validate the final result. The research further suggests the optimal rental quantity and price interval. This paper is the first to study the operational mechanism of sharing the parcel locker between two distinct types of logistics service providers and to offer recommendations for industrial application.

Morphophysiological and biochemical responses of cotton (Gossypium barbadense L.) to nano zinc (ZnO-NPs) and Azospirillum sp. under water deficit stress conditions

Water deficit stress (WDS) is one of the most significant abiotic limiting factors in cultivated crops, including cotton (Gossypium barbadense L.). With global climate change and the destruction of ecological balance, the frequency and severity of drought events are increasing in many regions around the world. Therefore, the aim of the present work was to investigate the potential of using nano zinc particles (ZnO-NPs), plant growth-promoting rhizobacteria (Azospirillum sp.), and their combination to mitigate the negative impacts of WDS on cotton var. Giza 96. Extended irrigation intervals of 30 and 45 days led to considerable decreases in plant height, chlorophyll content, relative water content (RWC), yield, and cotton fiber quality compared to the optimal irrigation interval (every 15 days). When applied individually or in combination, ZnO-NPs and Azospirillum sp. can ameliorate the negative effects of WDS on cotton growth and productivity. Overall, the use of Azospirillum sp. and ZnO-NPs, either individually or in combination, has demonstrated their potential to enhance cotton growth and yield parameters (plant height, dry weight, leaf area, chlorophyll pigment, seed index, seed yield, and lint%) under prolonged irrigation intervals during 2021 and 2022 seasons. Antioxidant enzymes activity comprising catalase (CAT), peroxidase (POD), and polyphenol oxidase (PPO) and proline content were increased under drought stress. Cotton fiber quality parameters including length, strength, and micronaire were insignificantly affected under low irrigation rate. The combination of ZnO-NPs and Azospirillum sp. provided the most effective mitigation of WDS during both experimental seasons, outperforming the effects of individual applications.

Risk of colorectal neoplasia after removal of conventional adenomas and serrated polyps: a comprehensive evaluation of risk factors and surveillance use

BackgroundSurveillance colonoscopy after polyp removal is recommended to prevent subsequent colorectal cancer (CRC). It is known that advanced adenomas have a substantially higher risk than non-advanced ones, but optimal intervals...

Integrated Optimization of Quality Control and Maintenance Policies for Light-Emitting Devices

This research represents a comprehensive exploration of optimization policies pertaining to burn-in, quality control, and preventive maintenance for light-emitting devices. Utilizing a MATLAB-based computational model, we embarked on an in-depth investigation of the complex interplay and dependencies between these policies. Our findings illustrated that, with optimal burn-in time, cut-off threshold, and replacement interval, the total cost per usage duration was significantly reduced. Additionally, we investigated the system performance with and without quality control. It was observed that an increase in burn-in time corresponded to an increase in cost. The absence of quality control resulted in a similar pattern with the replacement period, where the cost remained constant after a certain replacement period. Conversely, the implementation of quality control indicated a direct relationship between burn-in time and cost, and an inverse relationship between replacement period and cost.

Study of the instability of the fixed points cells temperature of working standards: establish the optimal time intervals between the comparisons

The cells of fixed freezing points of metals are an important component of the working standards of temperature of the 0th category. These standards are intended to reproduce the unit of temperature kelvin in accordance with ITS-90, which is currently the main method in the temperature range from –189 to 1084 °C. The transfer of the temperature unit is carried out by comparing the cells of the working standards of temperature with the cells of the state secondary standard of temperature. In order to develop recommendations for increasing the interval between comparisons and changing the requirements of the State Verification Scheme, one of the most important characteristics of the cells of the working standards of temperature was studied – the instability of the cell temperature in the interval between comparisons. The paper summarizes the results of the study of 26 cells of fixed points of tin, zinc and aluminum, the service life of which was from 2 to 10 years, and the interval between comparisons was from 2 to 5 years. Possible causes of instability of the reproducible temperature are described and the uncertainty of comparisons of working standard cells with secondary standard cells is calculated. The analysis of the results showed that the instability values over a long period of time do not exceed 1.02 mK for tin cells, 1.75 mK for zinc cells and 6.47 mK for aluminum cells. The most probable cause of deviation of the temperature of working standard cells from secondary standard cells is the different purity of the metals used. It is noted that the destruction of the quartz shell of the cell without mechanical action is unlikely, and the effect of pressure on the temperature of the reference point is absent. Based on the results of the comparative analysis, it was concluded that it is possible to increase the interval between comparisons for ampoules of tin and zinc to five years. Such an increase in the interval has a positive economic effect by reducing the cost of operating the state secondary temperature standard and transporting ampoules to the comparison site. For aluminum cells, an individual approach to setting the interval between comparisons is recommended based on the analysis of the results of previous comparisons. The materials published in this paper may be useful to specialists of metrological centers working with working temperature standards, as well as to all scientists studying the processes occurring in cells for measuring the solidification temperature of metals.

Demonstration and optimization of coherent Doppler wind LiDAR with low sampling resolution

A low sampling resolution scheme for coherent Doppler wind LiDAR (CDWL) is proposed. The CDWL offers advantages in precision and detection resolution but suffers from the requirement of high-speed data acquisition (DAQ) with high sampling resolution, such as 12- or 14-bit, which leads to an increase of the computational complexity and the system cost. The use of a DAQ system with lower sampling resolution can provide a solution to mitigate this problem. The feasibility of the proposed scheme is validated by simulations and experiments. The detection performance can be greatly affected by the quantization interval selected during sampling. It is shown that the optimal quantization interval exists and only depends on the carrier-to-noise ratio (CNR), and the optimal quantization intervals of several sampling resolutions are given at different CNRs. With the given optimal quantization configuration, the low sampling resolution data can be used for reliable wind field measurements. For long-distance detection with a CNR lower than −13dB, the CNR deterioration of 1-bit, 2-bit, 3-bit, and 4-bit signals can be as low as 2, 0.5, 0.2, and 0.1 dB.

Synergistic assessment of multi-scenario urban waterlogging through data-driven decoupling analysis in high-density urban areas: A case study in Shenzhen, China

Extreme meteorological events and rapid urbanization have led to serious urban flooding problems. Characterizing spatial variations in flooding susceptibility and elucidating its driving factors are essential for preventing damages from urban pluvial flooding. However, conventional methods, limited by spatial heterogeneity and the intricate mechanisms of urban flooding, frequently demonstrated a deficiency in precision when assessing flooding susceptibility in dense urban areas. Therefore, this study proposed a novel framework for an integrated assessment of urban flood susceptibility, based on a comprehensive cascade modeling chain consisting of XGBoost, SHapley Additive exPlanations (SHAP), and Partial Dependence Plots (PDP) in combination with K-means. It aimed to recognize the specific influence of urban morphology and the spatial patterns of flooding risk agglomeration under different rainfall scenarios in high-density urban areas. The XGBoost model demonstrated enhanced accuracy and robustness relative to other three benchmark models: RF, SVR, and BPDNN. This superiority was effectively validated during both training and independent testing in Shenzhen. The results indicated that urban 3D morphology characteristics were the dominant factors for waterlogging magnitude, which occupied 46.02 % of relative contribution. Through PDP analysis, multi-staged trends highlighted critical thresholds and interactions between significant indicators like building congestion degree (BCD) and floor area ratio (FAR). Specifically, optimal intervals like BCD between 0 and 0.075 coupled with FAR values between 0.5 and 1 have the potential to substantially mitigate flooding risks. These findings emphasize the need for strategic building configuration within urban planning frameworks. In terms of the spatial-temporal assessment, a significant aggregation effect of high-risk areas that prone to prolonged duration or high-intensity rainfall scenarios emerged in the old urban districts. The approach in the present study provides quantitative insights into waterlogging adaptation strategies for sustainable urban planning and design.