Search

This paper describes how Machine Learning and Robust Optimization techniques can greatly improve cash logistics operations. Specifically, we seek to optimize the logistics followed by the different branches of a given bank. Machine Learning is used to forecast cash demands for each of the branches, taking into account past demands and calendar effects. These demand predictions are forwarded to a Robust Optimization model, whose outputs are the cash transports that each branch should request. These transports guarantee that demand is fulfilled up to the desired confidence level, while also satisfying additional constraints arising in this particular domain. [ABSTRACT FROM AUTHOR]

Conventional Data Envelopment Analysis (DEA) assesses the performance of Decision-Making Units (DMUs) by measuring only one type of efficiency using definite data. However, due to the vastness of some industries, such as banks, calculating only one type of efficiency cannot confirm DMUs as efficient or inefficient. Also, real-world cases are often faced with uncertain data, and conventional DEA lacks the power to consider these uncertainties in evaluation. In this paper, we develop a multi-objective DEA model to calculate three types of efficiencies, including profitability, operational, and transactional for bank branches when there are uncertain data. First, we apply a modified DEA model which is capable of calculating the common weight of all inputs and outputs by solving only one linear programming model. Then, we employ a robust approach to handle the uncertainty in data. The uncertainty in data is described with discrete scenarios. Discrete scenarios require a set of possible values for each parameter with uncertain nature. Finally, we apply a fuzzy programming method to convert the proposed multi-objective model into a single-objective one. Our main goal is to calculate three types of efficiencies for bank branches under four different scenarios. To validate the accuracy of the proposed model, a real case of 45 Agriculture bank branches located in West Azerbaijan in Iran is examined. The results show that the proposed model can produce accurate results under different scenarios. We also perform a comparative analysis on each efficiency aspect to specify the benchmark branches and also inefficient branches. Comparative analysis can help managers recognize where improvement should be prioritized. [ABSTRACT FROM AUTHOR]

• A DEA framework for units in ordered hierarchies with shared functions is studied. • The proposed DEA framework controls comparability through a flexible constraint. • A procedure for obtaining economically feasible efficient targets is developed. • A case study of a branch network with three different branch categories is provided. The paper addresses the question of ensuring comparability in data envelopment analysis (DEA) in situations when units are organized in an ordered hierarchy with functions shared at different levels. In such a case, although units may have identical input-output sets that they put to use in similar production, they are not an ideally homogeneous group and their comparability in a benchmarking context is limited. The paper proposes to control explicitly the degree of comparability by a fairly flexible comparability constraint in order to obtain more informative technical efficiency scores and economically feasible targets. Furthermore, the paper develops a methodology to identify closest targets under the comparability constraint that are more attainable for inefficient units than traditional targets. These ideas are demonstrated in a case study located in the area of bank branch performance assessment from which the motivation of the paper sprouted. The case study shows for three hierarchical branch categories of a Slovak commercial bank that the comparability constraint renders closest targets more apposite, but they depend on how slacks are handled, i.e. whether they are summarized by a normalized sum or by means of a slacks-based measure. [ABSTRACT FROM AUTHOR]

This paper proposes a multi-step procedure that integrates robust methods, clustering analysis and data envelopment analysis (DEA) to identify bank branch managerial clusters and to study efficiency performance. By applying robust techniques based on principal component analysis, we look for (1) the detection of influential branches, i.e., exhibiting extreme operating behaviors, and (2) the clustering of branches based on operating characteristics. Our premise is that influential branches affect both the clustering and the determination of efficiency performance. The application of the procedure yields various aggregate influential-based branch profiles along with cluster profiles. These aggregate profiles provide valuable insights on the determinants of branch efficiency performance and operating patterns. Using the profiles as contextual information, DEA input-oriented slack-based models are applied to study branch efficiency performance from meta-frontier and cluster-frontier perspectives. Branch performance is characterized in terms of influential-based and cluster profiles, and efficiency designations. This allows for the understanding of how efficiency and peer selection are affected by influential branches, and how the profiles can be used to inform design decisions. [ABSTRACT FROM AUTHOR]

Abstract: This paper aims to explore the true managerial efficiencies of the branches of a case bank in Taiwan. With 123 branches of the case bank comprising the sample, the study finds that, after the adjustment of environmental factors and statistical noise, managerial efficiency values from a three-stage data envelopment analysis (DEA) varies significantly from the traditional DEA model. This finding suggests that environmental variables have significant effect on branch efficiency. Moreover, scale inefficiency is the major cause of operating inefficiency in the case bank, and most branches are operating at the stage of increasing return to scale. With regards the branches’ business scope, those that operate loan and wealth management services have better managerial efficiency than those that focus on wealth management only. In terms of deposit amount, branches with a higher deposit amount generate better managerial efficiency. Finally, the results for regional location show no significant effect on branches’ managerial efficiency in Taiwan. [Copyright &y& Elsevier]

• We propose a two-stage hybrid model that integrates ANN with DEA. • We position the problem in view of Design Science Research Methodology. • We study the model's ability to classify bank branches into predefined efficiency classes. • Bank branches may learn not only from best performers but also from better performers. • Results indicate a satisfactory classification ability especially for efficient bank branches. The quest for best practices may lead to an increased risk of poor decision-making, especially when aiming to attain best practice levels reveals that efforts are beyond the organization's present capabilities. This situation is commonly known as the "best practice trap". Motivated by such observation, the purpose of the present paper is to develop a practical methodology to support better practice benchmarking, with an application to the banking sector. In this sense, we develop a two-stage hybrid model that employs Artificial Neural Network (ANN) via integration with Data Envelopment Analysis (DEA), which is used as a preprocessor, to investigate the ability of the DEA-ANN approach to classify the sampled branches of a Greek bank into predefined efficiency classes. ANN is integrated with a family of radial and non-radial DEA models. This combined approach effectively captures the information contained in the characteristics of the sampled branches, and subsequently demonstrates a satisfactory classification ability especially for the efficient branches. Our prediction results are presented using four performance measures (hit rates): percent success rate of classifying a bank branch's performance exactly or within one class of its actual performance, as well as just one class above the actual class and just one class below the actual class. The proposed modeling approach integrates the DEA context with ANN and advances benchmarking practices to enhance the decision-making process for efficiency improvement. [ABSTRACT FROM AUTHOR]

• We propose a novel multi-criteria data envelopment analysis model. • The proposed two-stage DEA is first considered in the presence of ratio data. • Three states are considered (black-box, free-link, and fix-link). • A multi-criteria structure is used to demonstrate the validity of the models proposed. • An illustrative example on the efficiency of 60 sets of bank branches in Iran is provided. Data Envelopment Analysis (DEA) is a well-known technique for assessing efficiency levels of decision-making units (DMUs). Very often, available data may be expressed as ratios and, in such cases, traditional DEA models cannot be applied as long as biased efficiency results are produced, yielding the issues of efficiency underestimation and pseudo-inefficiency. In this paper, a novel two-stage MCDEA-R model to handle ratio data is developed observing three distinct assumptions – black-box, free-link, and fixed-link – offering a multi-criteria decision making (MCDM) perspective to the efficiency assessment problem in productive networks. While the proposed models are tested by evaluating the efficiency levels of a set of 30 bank branches in Iran, their distinctive features are highlighted in terms of previous literature to model ratio data under network structures. Precisely, there were not only gains in terms of mitigating pseudo-inefficiency and lack of discrimination power of weights issues, but there were also actual gains in terms of efficiency reliability as measured by information entropy. [ABSTRACT FROM AUTHOR]