Annals of Operations Research最新文献

In order to realize corporate sustainability, it is crucial for firms to embed environmental, social and governance (ESG) concerns into their business practices when embracing corporate social responsibility (CSR). The evaluation of business performance thus far provides limited support for the promotion of corporate sustainability since it rarely takes the nonfinancial information of ESG into account. Although much research has been carried out on the association between CSR and business performance, little attention has been given to incorporating CSR within the production system from an efficiency perspective. Using the well-known two-stage (profitability–marketability) data envelopment analysis (DEA) as a backdrop, this paper presents an extended framework that incorporates CSR, profitability and marketability into a parallel-series (mixed) structure of network DEA. A directional network slacks-based measure model, which allows the presence of shared resources and nonpositive data, is formulated to better evaluate firms’ business performance. The proposed model is illustrated with an empirical application to overall and three divisional efficiencies of 71 firms that were awarded Top 100 Excellence in CSR in Taiwan. The empirical results indicate that most sampled firms do not achieve full efficiency, highlighting considerable room for improvement. Among the dimensions of profitability, CSR, and marketability, CSR efficiency is relatively the lowest, yet it shows a strong positive association with overall efficiency, underscoring its crucial role in promoting sustainable business operations. In contrast, a negative relationship between profitability and CSR efficiencies suggests potential trade-offs between financial performance and social responsibility. The proposed network DEA framework enables firms to assess their competitiveness across these dimensions and set appropriate improvement targets, thereby supporting more sustainable business practices. These findings also provide valuable insights for investors seeking to make informed, ESG-oriented investment decisions.

Multi-class labelling in the absence of ground truth is a known hard problem in the computational intelligence paradigms. This problem is amplified in the case of e-commerce due to both high volume and high velocity of information. Specifically, it is hard to find labels for mass online reviews, rendering it unsuitable for supervised learning. Till date, the most sought solution is manual labelling, which remains a labour-intensive and time-consuming task. The purpose of this study is to develop an end-to-end approach for identifying the sources of quality-stimulated customer dissatisfaction and automatically assigning them in the context of e-commerce. The above objective is achieved by using a novel ensemble-based semi supervised pseudo-labelling technique on a large corpus of Amazon.com reviews. As a first step, a subset is manually labelled, followed by an ensemble approach of retaining commonly labelled (pseudo) class to iteratively label the entire dataset. We then apply Large Language Models (LLMs) and Deep Learning (DL) architectures on the (pseudo) labelled data to accomplish a multi-class classification problem. We contrast and showcase statistically significant improvement to the baseline machine learning models, where the pre-trained transformer models demonstrate best performance. Our approach proposes a roadmap to streamline automatically identifying sources of quality-related dissatisfaction in e-commerce channels using an amalgamation of ensemble and sophisticated computational techniques. We believe that our approach, if adopted, can bolster grievance redressal for online customers.

In various popular group decision making problems, e.g., proposal funding review, decision makers are required to provide both score values and sorting labels for alternatives. In this study, this kind of preference structure is formally defined as preference-sorting structure, which is a hybrid preference representation system extending the cardinal preference (score) model by combining multiple ordered categories in the process of preference modeling. In some cases, due to time and capacity limitations, the alternatives evaluated by a decision maker may constitute only a subset of the total alternatives. Several appealing axioms are introduced to characterize an “ideal” distance functions for preference-sorting structures in both complete and incomplete scenarios, respectively. Each set of axioms imply the existence of a unique distance function for complete and incomplete scenarios where the distance function for incomplete preference-sorting structures generalizes that for complete structures. Furthermore, based on the unique distance function for incomplete preference-sorting structures, an axiomatic distance approach with desirable properties is developed to aggregate individual incomplete preference-sorting structures. Through theoretical proof and simulation analysis, it is demonstrated that the group preference-sorting structure derived from the axiomatic distance approach exhibits a strong correlation with both the median score and median sorting.

Restless multi-armed bandits (RMAB) are a popular framework for modeling resource allocation and scheduling problems arising in various applications. Such applications can be modeled as Markov decision processes (MDP), but optimal or sub-optimal solution through dynamic programming suffer from high complexity. RMAB provides a heuristic solution, where the solution complexity scales linearly with the number of alternatives. However, these heuristic solutions are derived under the assumption that the model of all arms are known perfectly. In this paper, we consider RMAB with uncertainty in the rewards and dynamics of the arms. In such a setting, using a robust MDP solution is not possible due to high computational complexity. So, we consider a certainty equivalence approach and bound the additional loss in performance due to model inaccuracy. Our bounds are directly in terms of the model uncertainty of each arm and we illustrate their use via examples.

This editorial introduces the special issue Complexity, Nonlinearity and High Frequency Financial Data Modeling: Lessons from Computational Approaches in Annals of Operations Research, which brings together 19 contributions exploring advanced methods and applications in the analysis of financial markets. The collected works reflect the growing importance of complexity and nonlinear dynamics in understanding modern financial systems, marked by high volatility, interdependence, and structural shifts. The papers are organized thematically into five main areas: (i) complexity and nonlinearity in financial markets, (ii) advanced forecasting and econometric modeling, (iii) network theory, causality, and information flows, (iv) banking, credit risk, and economic growth, and (v) continuous-time and structural model reviews. There is an additional section on methodological innovations, which include time–frequency and multi-scale analysis, recent developments of nonlinear and regime-switching models, machine learning, and complex network approaches. A heartfelt tribute is dedicated to the late Marco Tucci, co-editor of this special issue, whose vision and scholarly contributions significantly shaped its content. Sadly, Marco passed away while we were in the process of compiling this special issue. The editorial concludes by highlighting common methodological threads, synthesizing key insights, and outlining promising avenues for future research in complexity-informed financial modeling.

Enterprises’ awareness and environmental crises induce the significance of corporate social responsibility (CSR), but demand disruption makes the members of logistics servicesupply chain (LSSC) cautious about profits. How do the CSR and demand disruption affect the optimal decisions of the LSSC? This study develops the two-echelon chain consisting of a logistics service integrator (LSI) and a functional logistics service provider (FLSP). Based on decentralized and centralized decision-making modes, the scenarios without and with demand disruption are studied. By comparing four analytical models, the results show that: (i) CSR could induce a lower price, and higher green innovation level and social welfare of the chain. A higher FLSP’s CSR ratio promotes her utility, but hurts the benefit and utility of LSI; (ii) as the disrupted demand increases, the level of service innovation, consumer surplus and corporate utilities ascend. It differs from logistics service price, which is determined by changing demand and CSR effort level; (iii) the channel utility and green innovation under demand disruption reach equilibrium in the centralized optimization, which greatly benefits the environment and logistics firms. The findings represent practical insights for mitigating the potential risk of demand volatility, and encourage logistics enterprises to proactively incorporate CSR.

In preference modelling, it is essential to determine the number of questions and their arrangements to ask from the decision maker. We focus on incomplete pairwise comparison matrices, and provide the optimal filling in patterns, which result in the closest weight vectors on average to the complete case for at most six alternatives and for all possible number of comparisons, when the underlying representing graph is connected. These results are obtained by extensive numerical simulations with large sample sizes. Many optimal filling structures resulted in optimal filling in sequences—one optimal case can be reached by adding a comparison to a previous one—which are presented on graph of graphs. The star graph is revealed to be optimal among spanning trees, while the optimal graphs are always close to bipartite ones. Regular graphs also correspond to optimal cases, furthermore regularity is important for all optimal graphs, as the degrees of different vertices are always as close to each other as possible. Besides applying optimal filling structures in given decision making problems, practitioners can utilize the optimal filling sequences in the cases, when the decision maker can abandon the problem at any period of the process (e.g., in online questionnaires).

This paper addresses decision-aiding problems involving multiple objectives and uncertain states of the world. Inspired by the capability approach, we focus on cases where a policy maker chooses an act that, combined with a state of the world, leads to a set of choices for citizens. While no preferential information is available to construct importance parameters for the criteria, we can obtain probabilities for the different states. To effectively support decision-aiding in this context, we propose two procedures that merge the potential set of choices for each state of the world taking into account their respective probabilities. Our procedures satisfy several fundamental and desirable propositions that characterize the outcomes.