perfomance management
ebrahim golzar; seyyed esmaeil najafi; seyyed ahmad edalatpanah; Amir Azizi
Abstract
Undesirable outputs are an integral part of production in various decision-making units, and to bring analyses closer to the real world, it is necessary to consider, undesirable outputs in performance evaluation research. In this paper, a new hybrid model for evaluating the efficiency of decision-making ...
Read More
Undesirable outputs are an integral part of production in various decision-making units, and to bring analyses closer to the real world, it is necessary to consider, undesirable outputs in performance evaluation research. In this paper, a new hybrid model for evaluating the efficiency of decision-making units in the oil industry is presented, which uses slack-based data envelopment analysis techniques and advanced machine learning algorithms. The proposed model specifically focuses on improving efficiency considering undesirable outputs and conditions of uncertainty. Three machine learning algorithms including artificial neural networks, support vector machines, and XGBoost are used to predict and improve the results of slack-based models. This study involves the evaluation of 37 decision-making units within the National Petroleum Products Distribution Company, and the results show a significant improvement in efficiency using predicted data compared to actual data. This research not only contributes to new perspectives in efficiency evaluation and improvement but also offers innovative hybrid methods to address challenges in operational management.
perfomance management
Esmaeil Keshavarz; abbas shoul; Ali Fallah Tafti
Abstract
Data Envelopment Analysis (DEA) is an approach based on mathematical programming for the relative evaluation of decision-making units treated as similar yet distinct production systems. In this approach, the performance of each unit is characterized by describing the transformation of specific inputs ...
Read More
Data Envelopment Analysis (DEA) is an approach based on mathematical programming for the relative evaluation of decision-making units treated as similar yet distinct production systems. In this approach, the performance of each unit is characterized by describing the transformation of specific inputs into specific outputs. Traditional DEA models assume that the role of each performance factor is clearly defined. However, in some real-world problems, certain factors might be identified as dual-role factors depending on the evaluation nature or the decision-makers' perspective. These factors can play the role of both input and output, or even be considered neutral in assessing the units' performance. In the current paper, to determine the status of dual-role factors and calculate the efficiency of DMUs, two new linear programming models, based on the concept of deviation in the efficiency constraint and a common set of weights, are suggested. The main advantages of the proposed models are significantly reducing the computations and iterations required to solve the model, and involving all DMUs to determine the role of factors. To assess the performance of the proposed models, a data set for the evaluation of eighteen suppliers in the presence of two inputs, three outputs, and two dual-role factors has been employed. The obtained results showed that, compared to other models, the proposed models are computationally more efficient, and the role determination and evaluation of the units, based on the obtained weights from these models, are better aligned with the expectations of decision-makers
modeling and simulation
Fereshteh Koushki
Abstract
It is inevitable for a manager to consider the performance effects of each component of a multi-stage financial equity capital. These components serve as inputs in the first stage to raise investments. The investments, as outputs of the first stage, become inputs for the second stage and are used in ...
Read More
It is inevitable for a manager to consider the performance effects of each component of a multi-stage financial equity capital. These components serve as inputs in the first stage to raise investments. The investments, as outputs of the first stage, become inputs for the second stage and are used in bank services, such as bank facilities, which are outputs of the second stage. Therefore, when evaluating bank performance, the connectivity between the stages must be considered; otherwise, efficiency may not be calculated correctly. Traditional methods often assess multi-stage systems as black boxes, neglecting the potential connectivity that may exist among the stages. We delve into the system and propose models to improve overall efficiency and the efficiency of each stage. Additionally, the continuity and relationships among stages introduce numerous variables and constraints to linear programming for evaluating the entire system. A centralized approach calculates the efficiency score of units simultaneously by solving only one linear programming problem, significantly reducing computational complexity. This approach, especially in large organizations, is commonly employed by central managers. In this paper, we introduce a centralized method for evaluating units with a multi-stage structure. We apply the proposed models to evaluate the efficiencies of bank branches and insurance companies, demonstrating the superiority of the improved network approach and centralized method in enhancing overall system efficiency. Bank branches typically have a two-stage structure, involving labor, physical capital, and other factors.IntroductionBank branches operate under the supervision of a central management team. The central manager, acting as the decision-maker, allocates resources such as labor and financial equity capital as inputs for these branches. The goal is to optimize the overall efficiency of the branches by minimizing the total consumption of resources while maximizing the desired outputs, such as security investments. A common approach to enhancing the performance of banks involves evaluating each branch separately. However, this method does not guarantee the minimization of total resource consumption and can be time-consuming. Since all bank branches are under the control of central management, the decision-maker can optimize the efficiency scores of branches by allocating resources to them simultaneously. This approach, known as centralized Data Envelopment Analysis (DEA), is particularly relevant when certain variables are controlled by a central authority, such as a Head Office, rather than individual unit managers. DEA is a mathematical programming technique used to assess the performance of homogeneous Decision Making Units (DMUs). However, in cases where DMUs have a network structure, such as banks, where the outputs of one division or sub-process serve as inputs for the next sub-process, traditional DEA models treat two-stage DMUs as black boxes and overlook potential connectivity among the stages. In our approach, we consider the internal activities within the system and propose a non-radial model to optimize multi-stage DMUs by taking into account the connectivity among the stages. Furthermore, in previous network DEA models, constraints related to intermediate activities were treated as inequalities, which, as we will demonstrate in this paper, can lead to contradictions in optimality. We address this issue by carefully considering the connectivity among stages. The presence of connectivity among stages introduces numerous variables and constraints to the corresponding model. This model, when used to measure the overall efficiency scores of all DMUs, would traditionally require solving as many problems as there are DMUs, which can be highly time-consuming. In our paper, we introduce a centralized approach that measures the efficiency scores of multi-stage structure DMUs by solving only one linear programming problem. We have applied these proposed models to evaluate bank branches and insurance companies. This approach provides a more comprehensive and efficient way to assess and improve the performance of multi-stage organizations like banks, taking into account the interconnected nature of their operations.MethodologyWe employ the Data Envelopment Analysis approach to evaluate systems with a multi-stage structure, often referred to as a network structure. Traditional DEA models treat two-stage DMUs as black boxes and overlook the potential for connectivity among these stages. In contrast, we delve into the internal activities of the system and propose a model that optimizes multi-stage DMUs by considering the interconnections among the stages. Moreover, in previous models designed to assess network systems, constraints related to intermediate activities were typically treated as inequalities, which could lead to inconsistencies in optimization. In our approach, we enhance these constraints associated with intermediate activities to ensure more robust optimization. Additionally, we apply a centralized approach to allocate resources to DMUs, allowing for the simultaneous optimization of the efficiency scores of all DMUs through the solution of a single linear programming problem. This centralized method streamlines resource allocation and improves the overall efficiency of the DMUs.ResultsWe evaluated 20 bank branches, treating them as 20 DMUs with a two-stage structure. In the first stage, inputs included paid interest, personnel costs, paid interest related to foreign currency transactions, and personnel costs related to foreign currency transactions. The first stage produced intermediate outputs in the form of raised funds and raised funds related to foreign currency transactions. In the second stage, the outputs consisted of loans and common incomes. Notably, some loans in the second stage might become non-performing, where borrowers are unable to make full or even partial repayments. To address this, we considered non-performing loans as undesirable or bad outputs and transformed them into inverse values to treat them as good outputs. To calculate the efficiency scores of the bank branches, we employed both our improved network model and the traditional DEA approach. Our network-based method revealed that many of the bank branches under evaluation were inefficient, in contrast to the traditional method, which inaccurately identified many of the bank branches as efficient. Subsequently, we extended our network method to a centralized case, significantly reducing computation time. The network-based assessment of bank branches took nearly 5 seconds, whereas solving the centralized model required only 0.1 second. In addition to evaluating bank branches, we applied our methods to assess insurance companies. The results demonstrated that our model provided more accurate efficiency scores compared to previous network-based approaches.ConclusionIn multi-stage production systems, the production process comprises several stages. Banks, for example, operate with a network structure in which labor, physical capital, and financial equity capital serve as inputs in the first stage to generate deposits as intermediate outputs. In the second stage, these banks utilize the deposits obtained from the first stage to create loans and security investments. We have introduced models to assess the efficiency of each stage, whether it's the first, intermediate, or final stage, individually. Additionally, we have developed a non-radial SBM model designed for evaluating DMUs with multi-stage structures. The Centralized DEA approach is a valuable method for central managers, particularly in large organizations like bank branches, to allocate resources effectively. We have extended our network-based method to a centralized approach, allowing us to calculate efficiency scores by solving just one linear programming problem. The results obtained from applying our proposed models to evaluate bank branches and insurance companies, both exhibiting network structures as DMUs, demonstrate the superiority of the network centralized approach over previous models.
Akbar Moradi; Alireza Amirteimoori; Sohrab Kordrostami; Mohsen Vaez-Ghasemi
Abstract
Data Envelopment Analysis (DEA) is an estimator. This estimator tries to assess a relationship between multiple inputs and multiple outputs, and an identified technology. In traditional DEA models, firms are classified into two divisions, efficient and inefficient. Efficient firms are considered as a ...
Read More
Data Envelopment Analysis (DEA) is an estimator. This estimator tries to assess a relationship between multiple inputs and multiple outputs, and an identified technology. In traditional DEA models, firms are classified into two divisions, efficient and inefficient. Efficient firms are considered as a reference for inefficient firms. In traditional DEA models, the efficiency improvement has been inspected for inefficient firms and efficient firms are assumed to be unchanged. Since the estimated technology is rationally smaller than the real technology or in other words, the estimated technology is always the subset of the true technology, we can expand it a little. Thus, we can improve efficient firms. This is done by creating some virtual DMUs. In this paper, an algorithm is proposed to expand the Production Possibility Set (PPS) and to improve efficient firms. To illustrate the proposed approach, numerical and applied examples are provided. The results are explained and discussed.
Davood Gharakhani; Abbas Toloie Eshlaghy; Kiamars Fathi Hafshejani; Farhad Hosseinzadeh Lotfi; Reza Kiani Mavi
Abstract
Data Envelopment Analysis (DEA) is a powerful analytical technique for measuring the relative efficiency for a set of Decision Making Units (DMUs) based on their inputs and outputs. There are weaknesses in conventional models DEA. Most important of which is the weight shift input and output which makes ...
Read More
Data Envelopment Analysis (DEA) is a powerful analytical technique for measuring the relative efficiency for a set of Decision Making Units (DMUs) based on their inputs and outputs. There are weaknesses in conventional models DEA. Most important of which is the weight shift input and output which makes the efficiency of Decision Making Units with different weights measured. A characteristic of Traditional DEA models is that it allows DMUs to measure their maximum efficiency score with the most favorable weights. As well as the conventional DEA models are not focused network of evaluation units. In this paper we propose to correct the weaknesses the common set of weights (CSW) in network DEA model based on the Goal programming approach. To test the effectiveness of the proposed model and solve real data is used by insurance companies active in Qazvin province. The model presented in this paper units decide on a similar scale with a set of weights for neutral evaluation is common. Proposed approach helps policy makers to better understand the strengths and weaknesses of DMUs and try to promote the strengths and remove weaknesses to improve the efficiency and ranking of given DMUs.
Akram Oveysiomran
Abstract
Input and output selection in Data Envelopment Analysis (DEA) has many important. In this research, inputs and outputs of reginal power companies are selected with artifitial neural network. The application of neural network in the selection of inputs and outputs of reginal power companies is not a precedent ...
Read More
Input and output selection in Data Envelopment Analysis (DEA) has many important. In this research, inputs and outputs of reginal power companies are selected with artifitial neural network. The application of neural network in the selection of inputs and outputs of reginal power companies is not a precedent in the literature and it is considered the main advantage of the proposed method. In order to train two layers MLP neural network, after presenting of error resilience, learning method was used. After neural network training, neural network performance is examined by using the test set. RMSE value for 15 test set equals 0/0269 which reflects the high accuracy of training network. The Sensitivity Analysis of the studied parameters which are the same inputs and outputs of Data Envelopment Analysis, with ten percent increase of parameter, compared to the prior one was carried out and output relative error average for neural network parameters was calculated. Based on the output relative error average, inputs and outputs were determined. By comparing the efficiency scores of regional electricity companies before and after reducing the number of variables, it is noticed that the number of efficient companies during the above four periods decreased from 50 percent to 11 percent. Finally, the neural network application in inputs and outputs selection of the regional electricity companies was unprecedented in the literature and this is the main advantage of this method.
