Annals of Operations Research最新文献

{"title":"A stochastic analytic hierarchy process framework for decision making in intensive care unit prioritization","authors":"Ankit Kumar,&nbsp;Talari Ganesh","doi":"10.1007/s10479-026-07279-7","DOIUrl":"10.1007/s10479-026-07279-7","url":null,"abstract":"<div><p>Multi-criteria decision-making problems in critical decision environments often suffer from two persistent challenges: experts often provide imprecise judgments, and classical AHP cannot adequately propagate this uncertainty. Existing stochastic or fuzzy variants help but still rely on pointwise simplification, heavy simulation, or ignoring compositional constraints. This paper introduces the Stochastic Analytic Hierarchy Process (SAHP), a closed-form Bayesian framework that integrates Beta-PERT triplet elicitation with a log-ratio Gaussian model to produce uncertainty-aware priority weights and ranking probabilities. SAHP directly transforms min-mode-max expert judgments into moment-based likelihoods, performs inference on a reduced-dimension log-weight space, and yields analytically tractable posterior distributions. This leads to a fast, transparent, and operationally feasible pipeline for uncertainty-propagating AHP. The framework is applied to a real-world ICU prioritization problem using clinical data from three COVID-19 waves. Results demonstrate that SAHP captures temporal shifts in clinical importance, produces stable and interpretable priority rankings, and offers clear uncertainty diagnostics through posterior intervals and rank probabilities. Comparative analyses show full agreement with classical AHP rankings under consistent inputs, while sensitivity tests confirm exceptional robustness to perturbations in elicited weights. The proposed SAHP advances the state of probabilistic MCDM by unifying practitioner-friendly elicitation, statistical rigor, and computational efficiency. The method is broadly applicable to healthcare, disaster response, infrastructure allocation, and any decision environment requiring rapid, explainable, and uncertainty-aware prioritization.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"364 3","pages":"3039 - 3069"},"PeriodicalIF":4.1,"publicationDate":"2026-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148965481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pricing strategy in livestream selling with the presence of myopic and strategic consumers","authors":"Wenting Yang,&nbsp;Jiantong Zhang,&nbsp;Devika Kannan,&nbsp;Dan Li,&nbsp;Song Xu","doi":"10.1007/s10479-026-07142-9","DOIUrl":"10.1007/s10479-026-07142-9","url":null,"abstract":"<div><p>Following the development of e-commerce, the manufacturer can sell products via direct or livestream modes. To meet market demand, the manufacturer will strategically adjust product prices, and consumers weigh whether to make early- or later-stage purchases based on their needs. In the current market, we assume two types of consumers including myopic and strategic consumers. Meanwhile, the tolerance level of each consumer regarding whether to delay purchase is heterogeneous. In our work, we built a two-stage game model to analyze the impact of consumers’ strategic behavior and their patience to delay purchases on manufacturer’s choice of the stage to adopt the livestream selling strategy under three scenarios including the manufacturer chooses a direct format to sell products in both stages (NN), the manufacturer chooses a direct format in the first stage and a livestream format to sell products in the second stage (NL), the manufacturer chooses a livestream format in the first stage and a direct format to sell products in the second stage (LN). Our main findings show that when strategic consumers dominate the market and consumers are more willing to postpone their purchasing, the manufacturer should adopt a penetration pricing strategy under NL scenario; otherwise, a skimming pricing strategy can be adopted. Under LN scenario, the manufacturer should directly employ a skimming pricing strategy. In addition, when the discount factor and the proportion of myopic consumers are small, and the commission rate is high, the manufacturer is more inclined toward the NL scenario; otherwise, it prefers to the LN scenario. In the extended models, we discuss the influence of different parameters, including the changed popularities of live-streamer, commission rates, and pit fees for the profit of the manufacturer. Moreover, we explore how the manufacturer selects livestream selling in both stages, and compare it with three main scenarios by using numerical studies. The result shows that the LN scenario is optimal in most cases.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"417 - 455"},"PeriodicalIF":4.1,"publicationDate":"2026-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07142-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generalized binomial models for analytics and decision-making: capturing feedback dependencies","authors":"Zvi Drezner,&nbsp;Dawit Zerom","doi":"10.1007/s10479-026-07255-1","DOIUrl":"10.1007/s10479-026-07255-1","url":null,"abstract":"<div><p>This paper investigates and extends the Generalized binomial distribution (GBD) as a novel extension of the classical binomial distribution designed to model dynamic feedback dependencies commonly encountered in decision-making contexts. By adjusting success probabilities based on prior outcomes, the GBD addresses critical limitations of traditional binomial and beta-binomial models, such as underestimating variance and failing to capture complex behaviors like bi-modality. Through extensive simulation studies and real-world applications, including marketing optimization and stock performance analysis, we demonstrate the versatility of the GBD in capturing feedback-driven phenomena. We also provide interactive tools (dashboards) to facilitate parameter exploration and practical implementation. These findings establish GBD as a valuable analytical tool for decision-making in management and operations offering both enhanced interpretability and actionable insights while retaining the simplicity of classical binomial frameworks.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"83 - 104"},"PeriodicalIF":4.1,"publicationDate":"2026-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scenario tree reduction via Wasserstein barycenters","authors":"Daniel Mimouni,&nbsp;Paul Malisani,&nbsp;Jiamin Zhu,&nbsp;Welington de Oliveira","doi":"10.1007/s10479-026-07062-8","DOIUrl":"10.1007/s10479-026-07062-8","url":null,"abstract":"<div><p>Scenario tree reduction techniques are essential for achieving a balance between an accurate representation of uncertainties and computational complexity when solving multistage stochastic programming problems. In the realm of available techniques, the Kovacevic and Pichler algorithm (Ann. Oper. Res., 2015 Kovacevic and Pichler (2015)) stands out for employing the nested distance, a metric for comparing multistage scenario trees. However, dealing with large-scale scenario trees can lead to a prohibitive computational burden due to the algorithm’s requirement of solving several large-scale linear problems per iteration. This study concentrates on efficient approaches to solving such linear problems, recognizing that their solutions are Wasserstein barycenters of the tree nodes’ probabilities on a given stage. We leverage advanced optimal transport techniques to compute Wasserstein barycenters and significantly improve the computational performance of the Kovacevic and Pichler algorithm. Our boosted variants of this algorithm are benchmarked on several multistage scenario trees. Our experiments show that compared to the original scenario tree reduction algorithm, our variants can be eight times faster for reducing scenario trees with 8 stages, 78 125 scenarios, and 97 656 nodes.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"271 - 291"},"PeriodicalIF":4.1,"publicationDate":"2026-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07062-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Connectedness spillover matrices : a tool for diversification","authors":"Daniel González Cortés,&nbsp;Monomita Nandy,&nbsp;Suman Lodh","doi":"10.1007/s10479-026-07174-1","DOIUrl":"10.1007/s10479-026-07174-1","url":null,"abstract":"<div><p>This research analyzes the performance and interconnectedness of major global stock market indices and decentralized finance assets, specifically cryptocurrencies, over the period from 2015 to 2025. The study includes indices such as the S&amp;P 500 and Nasdaq Composite from the United States, the FTSE 100, DAX, and CAC 40 from Europe, and the Nikkei 225 from Japan, and two more indices from China and India representing different economic regions. Additionally, Bitcoin and Ethereum are included to assess the impact of decentralized finance on traditional financial indices and asset allocation strategies. By employing Artificial Intelligence algorithms like ConvLSTM, the research measures the dynamic asset allocation and volatility management through an interconnected spillover matrix. The findings reveal that integrating ConvLSTM enhances the understanding of the interconnectedness between cryptocurrencies and traditional assets, offering improved diversification opportunities due to their low correlation, decentralization, and inflation-hedge characteristics. The study’s results suggest that investors can make more informed decisions regarding dynamic asset allocation in high-volatility portfolios, providing indicators of rising systemic risk and market stress.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"179 - 195"},"PeriodicalIF":4.1,"publicationDate":"2026-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07174-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A covering tour-based inventory routing framework for humanitarian logistics","authors":"H. Cansın Uzgören Kazanç,&nbsp;Mehmet Soysal,&nbsp;Zehra B. Kanık,&nbsp;Leandro C. Coelho","doi":"10.1007/s10479-026-07260-4","DOIUrl":"10.1007/s10479-026-07260-4","url":null,"abstract":"<div><p>In post-disaster situations, swiftly delivering humanitarian assistance to victims amid chaos and uncertainty poses a significant challenge in practice. Furthermore, efficient distribution of restricted resources, effective inventory control, and optimal resource allocation remain imperative priorities for humanitarian organizations that strive to meet urgent needs under adverse conditions. This study proposes a two-echelon Covering Inventory Routing Problem (CIRP) that integrates the Inventory Routing Problem (IRP) and the Covering Tour Problem (CTP) to support decision-making in the distribution of medical kits in post-disaster humanitarian logistics. A scenario-based probabilistic Mixed-Integer Linear Programming (MILP) model is introduced to decrease costs while adequately addressing unpredictable demand. The applicability of the model was assessed through scenario analysis and a case study. In addition, a three-phase matheuristic algorithm is proposed to solve the CIRP. The results demonstrate that integrating IRP and CTP in a two-echelon structure improves both cost efficiency and the reach of aid delivery under uncertainty. The use of a static-dynamic inventory approach, together with coordinated routing, effectively minimizes emergency shipments and adapts to fluctuating demand, providing valuable support for decision-making in real-time humanitarian contexts. The three-phase matheuristic achieved cost reductions of over 70% relative to the model’s incumbent solution within the first hour on large-scale instances, highlighting its practical use in accelerating decision-making amid post-disaster uncertainty.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"357 - 397"},"PeriodicalIF":4.1,"publicationDate":"2026-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07260-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The optimal asset allocation of national insurance funds","authors":"Natalia Gitelson,&nbsp;Eran Manes","doi":"10.1007/s10479-026-07257-z","DOIUrl":"10.1007/s10479-026-07257-z","url":null,"abstract":"<div><p>Survivability of the Social Security reserve is essential for the continuity of the welfare system. We take a novel approach to studying an optimal investment policy for the Social Security reserve that maximizes its survivability prospects. Applying a battery of optimization techniques, we find (i) High susceptibility to initial conditions: Countries that start with a low reserve-to-outflow ratio must invest heavily in stocks, whereas relatively rich economies can afford a more conservative portfolio. (ii) Survivability generally increases with the degree of exposure to stocks. (iii) A tradeoff exists between survivability in the short and long run. Societies that favor long-term survivability invest more in stocks, whereas countries that prefer avoiding default in the short run invest in bonds. Simulations show, however, that the discount rate must be unusually high for bonds to dominate stocks. Applying our results to the U.S., we find that a shift from the current investment regime to the suggested optimum improves welfare by 138 percent.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"149 - 177"},"PeriodicalIF":4.1,"publicationDate":"2026-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A p-median based approach to constrained clustering","authors":"Zhifeng Guo,&nbsp;Jesse R. O’Hanley,&nbsp;Stuart Gibson,&nbsp;Maria Paola Scaparra","doi":"10.1007/s10479-026-07259-x","DOIUrl":"10.1007/s10479-026-07259-x","url":null,"abstract":"<div><p>Constrained clustering is a semi-supervised method for dividing multi-attribute datasets into groups of similar elements that incorporates restrictions on how data points should or should not be clustered together. In this study, we investigate the application of the p-median model coupled with instance-level constraints, specifically must-link and cannot-link constraint, for performing constrained clustering. A Lagrangian relaxation procedure is proposed to efficiently solve large problem instances. To further speed up solution times, a simple but highly effective variable reduction technique is devised for identifying a small set of potential cluster centers. We test our modeling framework and solution approach on a set of benchmark datasets and several real-world household electricity consumption datasets. We find that high-quality solutions with small optimality gaps can be obtained with moderate computational effort. In addition, analysis of the largest household electricity consumption dataset reveals that constrained p-median clusters have less extreme variability with respect to average energy usage, a more even spread of cluster sizes, and much less overlap between high- and low-income households within the same clusters compared to classic k-means clustering.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"362 1-3","pages":"197 - 218"},"PeriodicalIF":4.1,"publicationDate":"2026-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07259-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148394928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The 2-Shapley value for bicooperative games","authors":"Manuela Basallote,&nbsp;Hugo A. Galindo,&nbsp;Inés M. Gallego,&nbsp;Andrés Jiménez-Losada","doi":"10.1007/s10479-026-07245-3","DOIUrl":"10.1007/s10479-026-07245-3","url":null,"abstract":"<div><p>In this paper, we propose a new Shapley-type value (the 2-Shapley value) for bicooperative games. Unlike classical cooperative games, bicooperative games assign values to pairs of disjoint coalitions (bicoalitions) representing players who support change and those who defend the status quo, while the remaining players abstain. To develop this new value, we employ antimatroid theory not as a structural constraint on the games, but as a methodological bridge. Specifically, we demonstrate that any bicooperative coalition structure is isomorphic to a specific antimatroid, which we term the ‘double antimatroid’. Leveraging this mathematical connection, we propose a new Shapley value, provide an axiomatization that preserves its fundamental axioms, and establish its consistency with existing solution concepts. Finally, we illustrate our theoretical proposal and the calculation of this value through an example modeling a decision-making process. This work provides a robust mathematical foundation connecting both structures, opening new perspectives for analysis within bicooperative game theory.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"361 2","pages":"503 - 527"},"PeriodicalIF":4.5,"publicationDate":"2026-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10479-026-07245-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148060622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal VPPI strategy under Omega ratio with stochastic benchmark","authors":"Guohui Guan,&nbsp;Lin He,&nbsp;Zongxia Liang,&nbsp;Litian Zhang","doi":"10.1007/s10479-026-07271-1","DOIUrl":"10.1007/s10479-026-07271-1","url":null,"abstract":"<div><p>This paper studies a variable proportion portfolio insurance (VPPI) strategy. The objective is to determine the risk multiplier by maximizing the extended Omega ratio of the investor’s cushion, using a binary stochastic benchmark. When the stock index declines, investors aim to maintain the minimum guarantee. Conversely, when the stock index rises, investors seek to track some excess returns. The optimization problem involves the combination of a non-concave objective function with a stochastic benchmark, which is effectively solved based on the stochastic version of the concavification technique. We derive semi-analytical solutions for the optimal risk multiplier, and the value functions are categorized into three distinct cases. Intriguingly, the classification criteria are determined by the relationship between the optimal risky multiplier in Zieling et al. (2014) and the value of 1. Simulation results confirm the effectiveness of the VPPI strategy when applied to real market data calibrations.</p></div>","PeriodicalId":8215,"journal":{"name":"Annals of Operations Research","volume":"361 2","pages":"609 - 636"},"PeriodicalIF":4.5,"publicationDate":"2026-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148060621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}