Search

As one of the most popular techniques for performance evaluation, Data Envelopment Analysis (DEA) has been widely applied in many areas. However, the self-evaluation used in DEA leaves it open to much criticism. Moreover, most researchers have ignored the fact that reality abounds with uncertainty and have assumed that the data used for evaluation is deterministic and accurate. Both assumptions make it difficult to evaluate the efficiency of real-world production processes correctly and reasonably. In this paper, we propose a series of robust cross-efficiency (RCE) models based on robust optimization theory and cross-efficiency to deal with these problems. First of all, the proposed RCE models allow the conservatism level to be adjusted easily to suit the attitude of the decision-maker towards uncertainty. In addition, the RCE models have better discrimination power than the existing robust CCR models. We present two applications to demonstrate the effectiveness and stability of our models. [ABSTRACT FROM AUTHOR]

The pharmaceutical supply chain (PSC) is a highly important strategic issue in the healthcare domain because of its significant effect on human wellbeing and healthcare costs. This paper provides a mathematical optimization model to design a PSC by considering the drug's corruption. Given the importance of the sustainable development goals, three objectives are considered: economic, social, and environmental. The cost of fixed installation, transportation, holding, collection, distribution, recycling and disposal is reduced by the economic objective. A reduction in greenhouse gas emissions, water use, and nonrenewable energy consumption is achieved by the environmental objective. Based on the social objective, the number of employment opportunities, economic growth, job satisfaction, and national self-sufficiency are all increased, while the number of days lost due to work-related accidents decreases. To achieve the weight of elements in environmental and social objectives, the fuzzy best–worst method is used. Due to the inherent uncertainty in the PSC, its main parameters are referred to as unpredictable, which calls for a robust optimization approach to eliminate the model parameters' uncertainty. A fuzzy programming approach is used to solve the developed model. Numerical evidence drawn from a practical case shows that the suggested model is efficacious and valid. We suggest that the priorities and challenges of sustainable growth contribute to the determination of the PSC's managerial and operational characteristics. [ABSTRACT FROM AUTHOR]

The training phase is the most crucial stage during the machine learning process. In the case of labeled data and supervised learning, machine learning entails minimizing the loss function under various constraints. We provide an innovative model for learning with numerous data sets, resulting from the application of multicriteria optimization techniques to existing deep learning algorithms. Data fitting is formulated as a multicriteria model in which each criterion measures the data fitting error on a specific data set. This is an optimization model involving a vector-valued function, and it has to be analyzed using the notion of Pareto efficiency. We present stability results for efficient solutions in the presence of input and output data perturbations. The multiple data set environment comes into play to eliminate the bias caused by the selection of a specific training set. To apply this concept, we present a scalarization strategy as well as numerical experiments in digit classification using MNIST data. [ABSTRACT FROM AUTHOR]

Efficiency plays a crucial role in portfolio optimization. This notion is formulated by means of stochastic optimization techniques. Very often this problem is subject to partial uncertainty or incomplete information on the probability distribution and on the preferences expressed by means of the utility function. In this case both the objective function and the underlying probability measure are not known with precision. To address this kind of issues, we propose to model the notion of incomplete information by means of set-valued analysis and, therefore, we propose two different extensions of the classical model. In the first one we rely on the notion of set-valued function while the second one utilizes the notion of set-valued probability. For both of them we investigate stability properties. These results are also linked to the notion of robustness of the aforementioned problem. Finally we apply the obtained results to portfolio theory and stochastic dominance. [ABSTRACT FROM AUTHOR]

We revisit in this article the Two-Fund Separation Theorem as a simple technique for the Mean–Variance optimization of large portfolios. The proposed approach is fast and scalable and provides equivalent results of commonly used ML techniques but, with computing time differences counted in hours (1 min vs. several hours). In the empirical application, we consider three geographic areas (China, US, and French stock markets) and show that the Two-Fund Separation Theorem holds exactly when no constraints are imposed and is approximately true with (realistic) positive constraints on weights. This technique is shown to be of interest to both scholars and practitioners involved in portfolio optimization tasks. [ABSTRACT FROM AUTHOR]