Operations Research Perspectives最新文献

The body of literature on goal programming (GP) approaches in modeling preferences and the satisfaction philosophy in multi-objective programming (MOP) decision-making processes is extensive. However, there has been little focus on how preferences change in relation to the decision-maker's (DM) behavior within this satisfaction philosophy, particularly in situations involving risk. To address this challenge, we propose introducing a behavior-type utility function into the GP model using the concept of a behavior function. This idea offers an innovative perspective for modeling DM's behavioral preferences in the imprecise GP approach by integrating a risk-aversion parameter specific to each objective. We then formulate a generalized behavioral-based GP approach for decision-making based on this new behavior-type utility function. To validate our proposed approach, we present an illustrative example of project selection in health service institutions, followed by a sensitivity analysis and comparisons with other approaches. The results demonstrate that DM's behavioral preferences significantly impact the decision-making process, and the proposed model provides more reasonable and convenient decisions for DMs with varying degrees of risk aversion.

In this paper, we analyze a perturbed formulation of bilevel optimization problems, which we refer to as $\delta$-perturbed formulation. The $\delta$-perturbed formulation allows to handle the lower level optimization problem efficiently when there are multiple lower level optimal solutions. By using an appropriate perturbation strategy for the optimistic or pessimistic formulation, one can ensure that the optimization problem at the lower level contains only a single (approximate) optimal solution for any given decision at the upper level. The optimistic or the pessimistic bilevel optimal solution can then be efficiently searched for by algorithms that rely on solving the lower level optimization problem multiple times during the solution search procedure. The $\delta$-perturbed formulation is arrived at by adding the upper level objective function to the lower level objective function after multiplying the upper level objective by a small positive/negative $\delta$. We provide a proof that the $\delta$-perturbed formulation is approximately equivalent to the original optimistic or pessimistic formulation and give an error bound for the approximation. We apply this scheme to a class of algorithms that attempts to solve optimistic and pessimistic variants of bilevel optimization problems by repeatedly solving the lower level optimization problem.

This paper presents a bi-level game model for pricing in a supply chain where the manufacturer (He) is the leader, and the retailer (She) is the follower. The leader decides on the wholesale price, and the follower decides on the selling price and selects seed nodes. The main idea of the model is that the marketing strategy used for promoting the product is focused on giving free samples to potential customers. Hence, the importance of analyzing a social network becomes evident. To maximize her profit, the retailer decides based on three factors: first, the leader's decision about wholesale price; second, the social network structure, which is critical for selecting the seed nodes; and third, people's valuation of the product. Therefore, a bi-level Mixed-Integer Nonlinear Programming model is developed to consider the social binds between potential customers. To solve this model, we employ a meta-heuristic algorithm. Finally, the effect of the model's parameters on decision variables and the objective functions is discussed. Based on the analysis and discussions, the production cost has a prominent impact on the players’ decisions and profits. Furthermore, instead of spending all the marketing budget on increasing seed nodes, it is suggested that they be spent on market research and improving good publicity. Moreover, deciding whether the players want to maximize the profit or market penetration is required before diving into decision-making.

Livestock production companies come under increasing responsibility to reduce their environmental impact, and thereby, the simultaneous decision-making of inventory replenishment and emissions reduction investment has become essential for ensuring sustainable development in the livestock farming business. This study investigates, for the first time, the best investment strategy for a livestock farming business under the carbon cap (CC) environmental legislation, taking into account both the edible and non-edible parts of slaughtering mature growing items (GIs) after procuring and feeding baby GIs. By fusing economic and environmental factors, this study aims to shed light on two crucial issues: (i) figuring out the appropriate level of investment needed for the farm to adhere to the CC environmental regulation; and (ii) evaluating the effect of the investment decision on the farm's expenses and emissions levels. To deal with these insights, a thorough analytical framework integrating mathematical modeling methodology, economic evaluation, and carbon accounting approaches is employed. By analyzing the interaction between the farm's emissions reduction investments and replenishment choices, the cost-effective investment level is determined that enables the farm to satisfy the carbon cap obligation while guaranteeing maximum operational efficiency. The results of this study have important ramifications for livestock farming businesses trying to make their way through the stringent CC emission law. The results indicate that in order to keep the business feasible when the cap of the CC guideline is low, the livestock-producing farm should give priority to investing in minimizing feed emissions and using cutting-edge manure treatment methods.

This paper studies a discrete-time portfolio optimization problem, wherein the underlying risky asset follows a Lévy GARCH model. Besides a Gaussian noise, the framework allows for various jump increments, including infinite-activity jumps. Using a dynamic programming approach and exploiting the affine nature of the model, we derive a single equation satisfied by the optimal strategy, and we show numerically that this equation leads to a unique solution in all special cases. In our numerical study, we focus on the impact of jumps and evaluate the difference to investors employing a Gaussian HN-GARCH model without jumps or a homoscedastic variant. We find that both jump-free models yield insignificant values for the wealth-equivalent loss when re-calibrated to simulated returns from the jump models. The low wealth-equivalent loss values remain consistent for modified parameters in the jump models, indicating extreme market situations. We therefore conclude, in support of practitioners’ preferences, that simpler models can successfully mimic the strategy and performance of discrete-time conditional heteroscedastic jump models.

As a resonance product between financial services and the upgrading of the green industry, green supply chain finance has garnered extensive attention in the process of ecological civilization construction. Effectively promoting the green transformation of small and medium-sized enterprises and achieving the "dual carbon" goals necessitate the avoidance of corporate green risks. However, the complex interdependence and information asymmetry among green supply chain finance enterprises result in data characteristics such as multi-source small samples and high-dimensional imbalance. To address these issues, this paper proposes a risk assessment model based on two-stage deep learning. In the first stage, we employ Generative Adversarial Network (GAN) to generate minority class default samples, and utilize Stacked Auto-Encoder (SAE) to extract data features with closed-form parameter calculation capability. In the second stage, the obtained features are input into a Deep Neural Network (DNN), and parameter learning and model optimization are conducted through joint training. Finally, to model low-order feature interactions, we integrate the Support Vector Machine (SVM) algorithm. The paper is grounded in the green innovation production of enterprises, collecting financial data of 176 upstream and downstream enterprises and corresponding core enterprise green indicators from 2013 to 2022. Experimental results demonstrate that GAN oversampling technique not only enhances the model's AUC metric but also significantly improves the F1 score. Compared with traditional deep learning methods, the proposed two-stage deep integration model effectively reduces training loss and exhibits superiority in identifying green supply chain finance risks.

Machine learning can be effectively used to generate models capable of representing the dynamic of production processes of small and medium-sized enterprises. These models enable the estimation of key performance indicators, and are often used for optimizing production processes. However, in most industrial applications, modeling and optimization of production processes are currently carried out as separate tasks, manually in a very costly and inefficient way. Automated machine learning tools and frameworks facilitate the path for deriving models, reducing modeling time and cost. However, optimization by exploiting production models is still in infancy. This work presents a methodology for integrating a fully automated procedure that embraces automated machine learning pipelines and a multi-objective optimization algorithm for improving the production processes, with special focus on small and medium-sized enterprises. This procedure is supported on embedding the generated models as objective functions of a reference point based non-dominated sorting genetic algorithm, resulting in preference-based Pareto-optimal parametrizations of the corresponding production processes. The methodology was implemented and validated using data from a manufacturing production process of a small manufacturing enterprise, generating highly accurate machine learning-based models for the analyzed indicators. Additionally, by applying the optimization step of the proposed methodology it was possible to increase the productivity of the manufacturing process by 3.19 % and reduce its defect rate by 2.15 %, outperforming the results obtained with traditional trial and error method focused on productivity alone.

We present a framework for multi-objective optimization where the classical mean–variance portfolio model is extended to integrate the environmental, social and governance (ESG) criteria on the same playing field as risk and return and, at the same time, to reflect the investors’ preferences in the optimal portfolio allocation. To obtain the three–dimensional Pareto front, we apply an efficient multi-objective genetic algorithm, which is based on the concept of $\varepsilon$-dominance. We next address the issue of how to incorporate investors’ preferences to express the relative importance of each objective through a robust weighting scheme in a multicriteria ranking framework. The new proposal has been applied to real data to find optimal portfolios of socially responsible investment funds, and the main conclusion from the empirical tests is that it is possible to provide the investors with a robust solution in the mean–variance–ESG surface according to their preferences.