Applied Soft Computing最新文献

{"title":"Distribution and expression aware retrospective learning for single-cell long-tailed class-incremental annotation","authors":"Tianhao Li ,&nbsp;Zixuan Wang ,&nbsp;Chenpeng Wu ,&nbsp;Zhengxiao Huang ,&nbsp;Yixin Xiang ,&nbsp;Yuhang Liu ,&nbsp;Quan Zou ,&nbsp;Naifeng Wen ,&nbsp;Yongqing Zhang","doi":"10.1016/j.asoc.2026.114745","DOIUrl":"10.1016/j.asoc.2026.114745","url":null,"abstract":"<div><div>Single-cell annotation, which identifies specific cell types in biological tissues, is a cornerstone of targeted therapy in precision medicine. However, the lack of incremental annotation capabilities in existing methods limits their adaptability to dynamic immune environment changes and broader applications. Additionally, the high-dimensional sparsity and long-tail distribution of single-cell data hinder practical long-tail incremental annotation. This paper introduces a single-cell incremental annotation framework that combines a distribution-aware diffusion model with an expression-aware knowledge distillation architecture to address these challenges. The distribution-aware module retrospects and replays historical data distributions, enabling precise generation of previously annotated cell types and mitigating catastrophic forgetting. The expression-aware module aligns gene representations through a multi-perspective attention mechanism, enhancing sensitivity to novel cell types while retaining knowledge of dominant categories. A fuzzy incremental guidance mechanism with uncertainty constraints further reduces the adverse effects of long-tail distributions, ensuring more robust annotation performance. Experimental results validate the effectiveness of this framework, demonstrating significant improvements in multi-session incremental annotation accuracy. In addition, this framework enhances the biological and clinical relevance of incremental annotation, enabling continuous integration of new datasets for improved discovery of rare cell types and disease progression analysis. This approach offers a scalable and generalizable solution to the challenges of long-tail incremental annotation in single-cell data.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114745"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledge-guided course recommendation in MOOCs with cross-graph and forgetting-aware cross-attention","authors":"Yangtao Zhou ,&nbsp;Qingshan Li ,&nbsp;Ruoyu Li ,&nbsp;Hua Chu ,&nbsp;Jianan Li ,&nbsp;Feifei Zhu ,&nbsp;Xiangming Li ,&nbsp;Chengyu Feng ,&nbsp;Wanqiang Yang","doi":"10.1016/j.asoc.2026.114768","DOIUrl":"10.1016/j.asoc.2026.114768","url":null,"abstract":"<div><div>Course recommendation (CR) plays a pivotal role in the e-learning field, aiming at tailoring course suggestions to learners based on their historical interaction behaviors. Existing CR methods predominantly utilize generic recommendation techniques to infer learners’ preferences from historical interactions. In course learning scenarios, learners’ interaction behaviors are not only preference-driven but also significantly influenced by their knowledge states. However, how to effectively capture the learners’ knowledge states in the CR task remains unexplored, and presents significant challenges due to two critical reasons. First, the knowledge states are intertwined with learners’ preferences, exhibiting mutual collaborative influences. Second, the knowledge states of learners dynamically decay over time due to natural forgetting processes. In this paper, we propose a novel knowledge-guided course recommendation method, named KnowCR, to capture the dynamic knowledge states of learners for enhancing course recommendations. Specifically, we propose a cross-graph collaborative encoding module that integrates the representations of learners’ preferences and knowledge states to exploit their collaborative influences. This module constructs two distinct graphs based on learner-course and learner-exercise interactions, and facilitates information integration via a cross-graph propagation mechanism. Moreover, we devise a forgetting-aware cross-attention encoding module that simulates the decay of short-term and long-term memories to capture the knowledge forgetting properties of learners. This module extracts the dynamic representations of learners by performing sequential encoding with a hierarchical knowledge forgetting mechanism on time-aligned interaction sequences. Extensive experiments on a real-world MOOC dataset demonstrate that our method significantly outperforms the state-of-the-art methods. The implementation codes and datasets are available at <span><span>https://github.com/KasISET/KnowCR</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114768"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards explainable image composition assessment: a dataset and a model","authors":"Quan Yuan ,&nbsp;Yipo Huang ,&nbsp;Pengfei Chen ,&nbsp;Leida Li","doi":"10.1016/j.asoc.2026.114784","DOIUrl":"10.1016/j.asoc.2026.114784","url":null,"abstract":"<div><div>Explainable Image Composition Assessment (Explainable ICA) provides quantitative scores and qualitative explanations. While recent research on ICA has demonstrated remarkable performance in regressing a scalar score for the overall composition perception, we still need a better understanding of how the image compositions, especially their influencing factors, relate to high-level semantics. However, the requirement for a high-quality explainable ICA dataset limits the broad real-world applications like intelligent photography and content recommendation. This deficiency is primarily because the manual annotation is costly, time-consuming, and struggles to capture the inherent aesthetic subjectivity of composition which heavily relies on expert knowledge. In this work, we take a step toward addressing this challenge through the use of multimodal large language models, and construct a large-scale composition description dataset based on the well-designed prompts in a Chain-of-Thought pattern, named CADB Comments. During its development, we ensured diversity across scenes, composition elements, and levels. This dataset comprises 9497 structured textual annotations. Leveraging this dataset, we propose Multimodal Image Composition Assessment (MICA) to bridge quantitative scoring and qualitative explanation, enabling simultaneous output of scores and descriptions. MICA generates descriptions, embeds composition knowledge into visual encoder, and extracts composition representation using the knowledge-enhanced encoder. A composition level contrastive alignment module strengthens correlation between representation and scores via contrastive learning. Extensive experiments on three datasets demonstrate MICA’s competitive performance in explanation, scoring, and classification. The dataset and code are available at <span><span>https://github.com/dylanqyuan/MICA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114784"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A key node detection method assisted by multi-layer surrogate model and multi-dimensional network features in complex networks","authors":"Hong Zhao ,&nbsp;Ling Tang ,&nbsp;Jia-Rui Li ,&nbsp;Zong-Gan Chen","doi":"10.1016/j.asoc.2026.114803","DOIUrl":"10.1016/j.asoc.2026.114803","url":null,"abstract":"<div><div>Identifying key nodes in complex networks is a critical task, as these nodes significantly shape network structure and functionality. Despite the progress of existing methods, two major challenges remain: high computational complexity in large-scale networks and the insufficient adaptability of evaluation models to diverse network structures. To address these challenges, we propose a key node detection method assisted by multi-layer surrogate model and multi-dimensional network features (KND-MSMNF). Our method introduces a multi-layer surrogate model-assisted network performance evaluation (MSNE) strategy, which integrates a Kriging model and the <em>K</em>-nearest neighbors (KNN) model within a differential evolutionary (DE) algorithm. By dynamically employing different surrogate models at various stages of population evolution, MSNE approximates the objective function, thereby circumventing computationally expensive evaluations and significantly improving the efficiency of network assessment and optimization. Furthermore, an adaptive multi-dimensional network feature search (AMNFS) strategy is proposed to evaluate node importance. This AMNFS strategy dynamically adjusts attribute weights and metrics based on network characteristics, leveraging a comprehensive set of features to avoid bias of single evaluation criteria and enhance adaptability. Additionally, a multimodal optimization-driven key node selection (MKNS) strategy assesses node impact from a global perspective by considering both network cost and overall connectivity. By identifying multiple global optima, MKNS improves the robustness of the solution set and provides flexible alternatives for decision-makers. Extensive experiments on diverse network datasets demonstrate that KND-MSMNF significantly outperforms existing methods in terms of key node detection accuracy (i.e., exceeds 96 %) while maintaining strong practicality and adaptability. Moreover, our method can simultaneously identify multiple combinations of key nodes. This enables the rapid localization of corresponding alternative key nodes in the event of a key node failure, thereby effectively mitigating the risk of network functional impairment caused by the failure of a single node.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114803"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An enhanced multi-criteria decision making framework for evaluating LLM-integrated smart product-service systems using spherical fuzzy rough numbers","authors":"Zhongkui Liu ,&nbsp;Danhui Zhang ,&nbsp;Qinghua Liu","doi":"10.1016/j.asoc.2026.114797","DOIUrl":"10.1016/j.asoc.2026.114797","url":null,"abstract":"<div><div>Integrating Large Language Model (LLM) into Smart Product-Service Systems (SPSS) introduces challenges in design evaluation due to inherent uncertainties and complex decision criteria. To address these challenges, we propose SFRN-LOPCOW-RAFSI, a novel multi-criteria decision making (MCDM) framework that combines spherical fuzzy rough numbers (SFRN), logarithmic percentage change-driven objective weighting (LOPCOW), and ranking of alternatives through functional mapping of criterion sub-intervals into a single interval (RAFSI). This study applies the proposed framework to a conceptual design evaluation case for a refrigerator service system, where four LLM-integrated service solutions are assessed based on 15 evaluation criteria. The methodology consists of three stages: (1) defining the problem structure, (2) computing the weight coefficients of the criteria using the SFRN-based LOPCOW method, and (3) ranking the design alternatives through the SFRN-based RAFSI approach. SFRN can represent a more extensive range of uncertainty and effectively handle uncertainty within the decision-making group, enhancing the reliability of the evaluation process. Sensitivity analysis and comparative analysis were conducted to validate the proposed method, confirming its robustness and effectiveness. This study's findings provide valuable insights for DMs in optimizing the integration of LLMs into SPSS design processes.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114797"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-objective decision-making control of cutter suction dredger based on multi-scale graph representation with Bayesian optimization","authors":"Yong Chen ,&nbsp;Qiubing Ren ,&nbsp;Mingchao Li ,&nbsp;Bolong Gao ,&nbsp;Yonghuang Xiang ,&nbsp;Zhongzhi Fu","doi":"10.1016/j.asoc.2026.114811","DOIUrl":"10.1016/j.asoc.2026.114811","url":null,"abstract":"<div><div>The operational parameters of cutter suction dredger (CSD) are frequently set based on empirical experience, leading to low productivity and high energy consumption. To address this issue, the present study proposes a Bayesian multi-objective optimization (MOO) framework integrating a multi-scale adaptive graph convolutional network (MAGCN) model and a multi-objective tree-structured Parzen estimator (MOTPE) algorithm. This framework aims to predict and optimize construction productivity and energy consumption performance while providing well-informed support for intelligent decision-making control. First, a hybrid feature selection method combining maximum information coefficient and Pearson correlation coefficient is applied to identify 6 key control parameters from 256 operational parameters. Second, an MAGCN model with multi-scale graph representation is proposed to establish the nonlinear mapping among geological conditions, operational parameters, construction productivity, and energy consumption. Third, the MOO framework with its mathematical formulation is constructed, and the Pareto optimal front for productivity and energy consumption is derived using the MOTPE-based Bayesian optimization algorithm. The optimal trade-off solution and corresponding operational parameters are determined by a weighted sum method. The proposed MOO framework is validated using operational data from the Tian Jing Hao CSD in the Pinglu Canal dredging project in China. The results show that the MAGCN model achieves high prediction accuracy for real-time productivity and energy consumption, with R² values of 0.942 and 0.979, respectively. The MOTPE-driven optimization reduces standard deviations of operational parameters by 21.24 % - 97.90 %, enhancing operational stability. The framework improves productivity by 3.79 % and reduces energy consumption by 3.53 %. This work provides real-time decision-making support for optimizing CSD operations.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"193 ","pages":"Article 114811"},"PeriodicalIF":6.6,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MFU-Depth: Modeling and fusing uncertainty for self-supervised multi-frame monocular depth estimation","authors":"Fuji Fu ,&nbsp;Jinfu Yang ,&nbsp;Jiaqi Ma","doi":"10.1016/j.asoc.2026.114705","DOIUrl":"10.1016/j.asoc.2026.114705","url":null,"abstract":"<div><div>Self-supervised monocular depth estimation, valued for its simplified configuration and excellent geometric perception, is gaining attention in autonomous driving and robotics. Mainstream self-supervised multi-frame methods, due to the lack of a unified multi-cue uncertainty representation mechanism, often struggle to effectively reflect depth prediction errors. To address this issue, we propose a self-supervised multi-frame monocular depth estimation method by Modeling and Fusing Uncertainty, termed MFU-Depth. (1) At the probability distribution level, we design an Uncertainty Estimation-based Single-frame Probability Prediction (UE-SPP) module and a Depth-Aware Multi-frame Probability Transformation (DA-MPT) module, achieving unified uncertainty modeling of single-frame and multi-frame cues in the form of probability distributions. Building on these, we establish a Relation Modeling-guided Probability Fusion (RM-PF) mechanism, which adaptively adjusts fusion weights through pixel-wise difference relation modeling, enabling the complementary integration of these two distributions. (2) At the probabilistic self-supervision level, an Information Entropy Uncertainty Loss (IEUL) is devised to further model the uncertainty of both single-frame and fused depth probability distributions, thereby mitigating supervision distortion caused by high-uncertainty pixels. Experimental results show that MFU-Depth achieves exceptional performance on multiple public datasets. Compared to baseline methods, MFU-Depth reduces the depth error by 15.3% and the uncertainty metric by 29.0%.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"191 ","pages":"Article 114705"},"PeriodicalIF":6.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MACSENet: A novel lightweight CNN with multi-scale atrous convolutions and attention mechanism for accurate lung cancer detection","authors":"Fazal Hadi ,&nbsp;Omair Bilal ,&nbsp;Sohaib Asif","doi":"10.1016/j.asoc.2026.114698","DOIUrl":"10.1016/j.asoc.2026.114698","url":null,"abstract":"<div><div>Lung cancer represents a critical global health challenge, characterized by high mortality rates and complex diagnostic processes. Identifying tumors in CT images manually is a time-intensive and complex task. While deep learning models can aid in the process, their complexity often leads to overfitting, and their single-branch, linear structures struggle to capture multi-scale features, ultimately reducing detection accuracy. This paper presents MACSENet, an innovative lightweight CNN developed to improve early detection and diagnostic accuracy using advanced deep learning techniques. The proposed model combines three key components: a standard CNN for basic feature extraction, a multiscale atrous convolutional module for capturing spatial features at different resolutions, and a squeeze-and-excitation (SE) block to refine channel-wise features through adaptive attention mechanisms. We designed the Multiscale Atrous Feature Extraction Block (MAFEB), which incorporates a multiscale atrous convolution module with varying dilation rates. This approach expands the receptive field without adding extra parameters, capturing intricate tumor details and providing a comprehensive contextual understanding to address the complex morphological variations in lung cancer. The SE block is applied after extracting features from the multiscale atrous convolution module, dynamically refining the feature maps by amplifying key diagnostic information and suppressing irrelevant details. To assess the broader applicability and generalizability of MACSENet, we evaluated it on the publicly available IQ-OTH/NCCD lung cancer dataset, which consists of 1097 CT images from 110 patients, encompassing benign, malignant, and normal cases. Additional experiments were carried out to evaluate the robustness of MACSENet using two other datasets: the Chest CT-Scan dataset, which consists of 1000 images across four classes, and the Lung histopathological images dataset, containing 15,000 images across three classes. The experimental results demonstrate that MACSENet outperforms both traditional deep CNNs and state-of-the-art methods in terms of accuracy and efficiency. Our approach achieved an outstanding accuracy of 99.55 % on the IQ-OTH/NCCD dataset, and 92 % and 99.97 % on the additional datasets while maintaining the smallest parameter size of 0.27 million among all competitors. Additionally, the model exhibited the smallest model size and FLOPS compared to other deep CNNs. Our proposed model is also compared with well-established baseline models including ResNet, DenseNet, and MobileNet series to demonstrate its superior performance in lung cancer detection. The proposed approach provides critical insights to support clinical decision-making for advancing patient care and treatment precision in lung cancer diagnostics.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"191 ","pages":"Article 114698"},"PeriodicalIF":6.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A two-stage solution incorporating large neighborhood search for the priority-aware 2D bin packing problem in furniture manufacturing","authors":"Jianming Wang ,&nbsp;Yang Xu ,&nbsp;Zhouwang Yang","doi":"10.1016/j.asoc.2026.114659","DOIUrl":"10.1016/j.asoc.2026.114659","url":null,"abstract":"<div><div>This paper explores the two-dimensional bin packing problem involving both rectangular and irregular shapes, with a focus on part priority, a critical factor in the furniture manufacturing and woodworking industries. Part priority is essential due to specific processing requirements and customer urgency. We propose a two-stage methodology aimed at minimizing both the number of bins containing priority parts and the total number of bins utilized. In the first stage, multiple parts are iteratively paired and approximated as rectangles to maximize the overall benefit of all pairings, diverging from traditional methods that focus on pairing two specific parts. The second stage introduces a rectangular packing module that incorporates two large-neighborhood search (LNS) algorithms. This module employs efficient operators that respect the priority objective, addressing the difficulty of large-scale priority problems. We evaluate the strengths and limitations of the rectangularization approach through experiments on irregular bin packing benchmarks and assess its applicability. Experiments on rectangular benchmark instances demonstrate the superiority of our approach in large-scale scenarios. Furthermore, tests on industrial data reveal that our method increases material utilization by 0.86% and reduces the number of priority bins by 1.71%, surpassing leading commercial software in both objectives. These results suggest that the proposed approach can be integrated into cutting software to provide practical and efficient solutions, thereby advancing intelligent manufacturing.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"191 ","pages":"Article 114659"},"PeriodicalIF":6.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A hybrid NSGA-II and delayed local search approach for home healthcare scheduling and routing optimization","authors":"Radhia Zaghdoud ,&nbsp;Issam Zidi ,&nbsp;Olfa Ben Rhaiem ,&nbsp;Salim El Khediri ,&nbsp;Khaled Mesghouni","doi":"10.1016/j.asoc.2026.114617","DOIUrl":"10.1016/j.asoc.2026.114617","url":null,"abstract":"<div><div>The home healthcare sector has emerged as a vital component of modern healthcare systems, addressing the needs of aging populations and individuals with disabilities. However, efficiently managing caregiver schedules and routing while balancing service time windows, skill compatibility, and operational costs remains a critical challenge. This study tackles a multi-objective Home Healthcare Scheduling and Routing Problem (HHCSRP) by optimizing three competing objectives: minimizing total travel time, ensuring equitable workload distribution among caregivers, and reducing waiting time between patient visits. We propose a novel two-stage hybrid algorithm that strategically integrates the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) with a tailored local search method. In the first stage, NSGA-II generates a diverse initial population over multiple generations, while the second stage embeds local search at mid-generations to refine solutions without premature convergence. Computational experiments using adapted Solomon’s Vehicle Routing Problem with Time Windows (VRPTW) benchmarks demonstrate the effectiveness of the algorithm. Our approach achieves superior hypervolume (0.57–0.86) and Pareto front diversity (<span><math><mo>|</mo><mi>F</mi><mi>P</mi><mo>|</mo></math></span> = 94.27 on average), outperforming existing methods in balancing efficiency, fairness, and scalability. The results highlight the robustness of the method in handling large-scale instances (100 patients), offering a scalable tool for real-world home healthcare logistics. This work advances multi-objective optimization in healthcare operations, providing actionable insights for administrators to harmonize patient satisfaction and operational efficiency.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"191 ","pages":"Article 114617"},"PeriodicalIF":6.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}