Annals of Operations Research最新文献

This study investigates the application of Multiple Criteria Decision Analysis (MCDA) methods for portfolio selection in the Vietnamese stock market using daily stock price data from the VN100 index spanning January 2015 to November 2023. Creating 150 criteria based on stock returns, volatility, and correlation, we employ five popular MCDA methods and four weighting methods to compare the performance of up to 500 portfolios. While MCDA methods may not effectively differentiate between stocks with higher and lower future returns, they consistently excel in selecting portfolios with superior risk-adjusted returns and lower drawdown compared to both the benchmark and the traditional Mean-Variance (MV) method. The PROMETHEE II (Preference Ranking Organization Method for Enrichment of Evaluations) method stands out as a robust performer, and the CRITIC (Criteria Importance Through Intercriteria Correlation) weighting method emerges as a reliable choice for constructing portfolios with favorable risk-adjusted returns. Moreover, the MCDA methods demonstrate potential computational efficiency over the tested MV implementation, enhancing their practicality. These findings highlight the practical utility of MCDA, particularly PROMETHEE II with CRITIC weighting, for navigating the complexities of portfolio optimization in dynamic emerging markets like Vietnam, offering a compelling alternative to traditional approaches.

When a disaster occurs, Humanitarian Logistics and Supply Chain Management (HLSCM) works to ensure the timely and appropriate distribution of relief goods. One is food distribution, one of the most needed essential supplies after a disaster. However, delays in the delivery of food logistics often occur, resulting in fatal consequences for disaster victims. In HLSCM, delays are referred to as risks, while the causes of delays are risk agents that must be identified. Identifying risk agents in distribution delays is crucial because it dramatically affects the entire HLSCM flow. This study proposes a hybrid literature search strategy combined with expert judgment to identify risk agents that cause food delivery delays in the HLSCM process. The results of this study indicate that there are 17 risk agents affecting food logistics delivery delays. Based on the mathematical aggregation of expert judgment results, the highest average values ​​are shown by three risk agents: “inventory quantity,” “road conditions,” and “communication problems.” This investigation fills the gap in the existing HLSCM literature by explicitly identifying and validating these risk factors. The hybrid literature search strategy combines a systematic literature review with expert judgment and provides a strong framework for future research. These findings offer practical insights for stakeholders to improve logistics efficiency and ensure timely aid delivery. While focusing on risk identification, this study lays the groundwork for future research to prioritize risk mitigation strategies and further refine HLSCM practices.

With the rapid development of e-commerce and platform economy, an increasing number of products exhibit network effects. Facing the network effects as well as system service congestion, the propensities of customers for interacting with other users are heterogeneous. In this paper, we investigate pricing in a two-stage sales model in the presence of network effects and customer heterogeneity, modeled as a single-server queueing model with congestion in each stage. In the two-stage sales system, initial sales are usually targeted at establishing the brand, while later sales are aimed at maximizing profit. In contrast to traditional sales, we find that the initial pricing decision under the influence of network effects can be optimized on either a subsidy or charging strategy. Specifically, our results indicate that the subsidy strategy is more advantageous under situations where either the second-stage price is relatively high or the strength of network effects is relatively low; otherwise, the charging strategy is better off. Further, it is revealed that strategic behavior of customers may drive prices up in the first stage, which in turn damages the firm’s revenue. Our results provide valuable insights that ignoring the impacts of network effects and strategic customer behavior may result in significant revenue loss to the firm.

The recycling and remanufacturing of electronic waste are critical for resource conservation and the sustainable development of both society and the environment. However, uncertainty about the quality of recycled products introduces risk aversion. To address this issue, the present study develops a differential decision-making model that distinguishes between manufacturers who consider, and those who ignore, the risk-averse behavior of recyclers. The analysis examines how recyclers’ risk aversion affects key factors in a closed-loop supply chain, including sales price, recovery rate, demand, and the profits and utilities of both manufacturers and recyclers. Our findings suggest that achieving optimal recovery rates, demand, and utility for both parties requires that carbon tax formulation aligns with recovery costs. Additionally, this tax serves as a critical benchmark for determining the repurchase price of waste products. When recyclers exhibit risk aversion, increasing this aversion results in higher product sales prices, but also leads to lower demand, recovery rates, and profits for both manufacturers and recyclers. In contrast, manufacturers who disregard recyclers’ risk aversion may lower product sales prices, thereby increasing recovery rates, demand, and recycler utility. However, acknowledging and addressing risk aversion has a positive effect on manufacturer profits and utility.

In the context of finding risk bounds for portfolios of risks, Puccetti and Rüschendorf (J Comput Appl Math 236(7):1833–1840, 2012) introduce the rearrangement algorithm (RA) as a tool for (optimally) rearranging matrices by permuting, in each step, the elements of a given column. The RA also has applications in finance and operations research. Bernard and McLeish (Asia-Pac J Oper Res 33(05):1650040, 2016) and Bernard et al. (J Risk Insur 84(3):923–959, 2017) show that, in principle, better results can be expected by permuting the rows of randomly chosen blocks of the matrix. They label such an algorithm the block rearrangement algorithm (BRA). Various versions of BRA exist, and they mainly differ with respect to the manner in which the blocks (i.e., the submatrices) are chosen in each step. In this paper, we aim to develop an improved version of BRA based on a dynamic choice of block sizes. That is, we seek to find the optimal sequence of block (submatrix) sizes. To achieve this, we refine the BRA by sampling the block size, (r_t,) at the t-th step from a Beta distribution with two parameters that evolve over the different steps. The proposed BRA Beta is designed to select large block sizes initially (similar to BRA Binomial), and then transition to smaller sizes (resembling the RA). A numerical study demonstrates that BRA Beta outperforms other variants of the RA available in the literature. For example, the improvement in variance reduction achieved by BRA Beta is double that of other BRA-based algorithms when the risks are heterogeneous and the portfolio is large (see Sect. 5).

Disasters impose huge human and financial losses and hinder humanitarian logistics. This study aims to find an efficient solution for humanitarian relief logistics in disasters by reducing time and cost. A sustainable-resilient multi-commodity, multi-depot, multi-objective, mixed-integer programming model is proposed for humanitarian relief logistics networks. The model considers network disruption, backup facilities, and rescue operations within standard relief time and golden time. The model utilizes a heterogeneous fleet of vehicles at different speeds, considering physical and mental health services at mild, moderate, and severe levels, and hygiene items such as diapers and sanitary pads in shelters. These components are included in the model to achieve Sustainable Development Goals and sustainability, the most important of which is Good Health and Well-being. After linearization, a case study is conducted to assess the validity of the model. Our proposed model is designed for a two-echelon location-allocation problem integrating pre- and post-disaster actions and solved via the augmented ε-constraint method in CPLEX/GAMS. It is observed that the model effectively allocates demands within a reasonable time and cost, finds facility locations, and activates backup facilities when warehouses are damaged due to disruption. The obtained results of the multi-objective model illustrate a trade-off between the objective functions, where a slight decrease in the response time leads to significantly higher costs. The sensitivity analysis demonstrates that an increase in demand leads to higher variable costs and longer response time but does not affect fixed costs. Furthermore, increasing the number of facility disruptions extends the relief time.

This study investigates the efficiency and performance of waste supply chain management across eight major South Korean cities, focusing on the interplay between input variables, inefficiency determinants, and waste processing outputs. Employing a multidisciplinary framework grounded in Resource-Based View, Environmental Justice Theory, and Systems Theory, the research utilizes Stochastic Frontier Analysis (SFA), Grey Incidence Analysis (GIA), and Artificial Neural Network (ANN) to evaluate the relative importance of various influencing factors. SFA estimate results highlight that budget and manpower productivity significantly contribute to efficiency, while disparities in budget allocation and outdated infrastructure contribute to inefficiencies. GIA underscores the dominance of commercial incineration and landfill performance, driven by strict industrial regulations and waste-to-energy initiatives. Conversely, commercial recycling and domestic landfill perform the worst. ANN reveals that budget productivity and manpower productivity have stronger and more impactful relationships with efficiency scores in cities like Seoul, Busan, and Incheon. On the inefficiency side, high facility installation costs, operation costs, and miscellaneous costs demonstrate significant negative impact on overall effectiveness across multiple cities.

In this paper, we re-examine 166 previously identified asset pricing characteristics and their ability to successfully predict stock returns. We use Explainable Artificial Intelligence to rank these return predictors based on their importance in various asset pricing model settings. Our findings suggest that ensemble and deep learning-based models have an advantage in providing generalized predictions across different return measures. Using SHapley Additive exPlanations, we also find that momentum and trading-based features possess higher predictive power in estimating asset returns. The long-short portfolio analysis reveals that key return predictors exhibit substantial economic significance, reflected in the large differences in out-of-sample (R^2). These findings remain robust across various models and persist even after controlling for characteristics-based predictors.