Stochastic Data Envelopment Analysis Research Articles

Inverse chance-constrained dynamic data envelopment analysis under natural and managerial disposability: Concerning renewable energy efficiency and potentials

The performance of dynamic systems with specific data has been examined in some data envelopment analysis (DEA) studies. However, in many real-life situations, dealing with dynamic systems becomes challenging because of random factors. So, in this paper, a chance-constrained dynamic DEA approach under managerial and natural disposability is first proposed to analyze the dynamic renewable energy efficacy of processes in the presence of random performance measures. Then, the potentials of some performance metrics are addressed for changes of others using the introduced inverse chance-constrained dynamic DEA models. The chance-constrained dynamic DEA techniques and their inverse dynamic stochastic problems are converted into linear problems. The introduced approaches are applied to probe the renewable energy efficiency and potentials of some OECD countries in a span of time. The results show stochastic dynamic DEA and inverse stochastic dynamic DEA provided are practical tools for making the best choices, when there is uncertainty.

Data Envelopment Analysis and Models in Healthcare Services: A Review

Data Envelopment Analysis (DEA) is a widely used method for measuring efficiency and performance in healthcare services. In this article, we will examine the role of DEA and the models used in the healthcare sector. DEA is a method used to measure the efficiency of units (such as hospitals, clinics) with multiple inputs and outputs. Essentially, data envelopment analysis conducts a performance analysis that shows how units utilize their existing resources and how they can optimize these resources. Units such as hospitals, clinics, and healthcare organizations must deliver more services with limited resources. DEA helps these organizations to use their resources most efficiently. DEA is used to increase operational efficiency, reduce costs, and improve service quality in healthcare services. There are various models available for data envelopment analysis. The most common ones include Data Envelopment Analysis (DEA) and Stochastic Data Envelopment Analysis (SDEA). While DEA is used for measuring efficiency that can be either constant or variable in scale, SDEA considers uncertainties and random effects. There are various advantages and disadvantages. Facilitates efficient use of resources, brings a data-driven approach to the decision-making process and objectively evaluates performance in healthcare services. Disadvantages are data deficiencies or poor quality may pose challenges in some cases and the complexity of the model may complicate the application and interpretation process. Data envelopment analysis in healthcare services is a powerful tool for measuring and improving the efficiency of organizations. However, proper use of data and careful application of the model are essential.

Assessment of carbon efficiency in wastewater treatment plants through Stochastic non-parametric data envelopment analysis (StoNED): Insights from Spanish facilities

Evaluating the carbon efficiency (CE) of wastewater treatment plants (WWTPs) is crucial for guiding these facilities towards carbon neutrality. Nevertheless, enhancing carbon performance ought not to be pursued to the detriment of pollutant removal efficiency. In this study, the CE of WWTPs is calculated as a composite index that integrates carbon emissions with the volume of pollutants removed from wastewater, including organic matter, suspended solids, nitrogen, and phosphorus. To achieve this, the Stochastic Non-parametric Envelopment of Data (StoNED) method is employed. This approach, distinct from the commonly used Data Envelopment Analysis (DEA), accounts for both empirical data and random variations in the estimation of CE, thereby enhancing the reliability of the CE evaluation. The study assesses the CE of 109 Spanish WWTPs, finding that none of them are fully carbon efficient. This involves that all plants have potential to reduce greenhouse gas emissions without sacrificing pollutant removal efficiency. The average CE of the WWTPs is 0.529, indicating a possible reduction in carbon emissions by approximately 0.076 kg of CO2 equivalent per cubic meter of wastewater treated. The analysis also reveals that neither the volume of wastewater treated nor the type of reactor used for secondary treatment has a significant impact on the CE of the facilities. The CE metric proposed in this study serves as an important decision-support tool for advancing towards the carbon neutrality of wastewater treatment processes. By providing a more comprehensive understanding of the environmental performance of WWTPs, it helps identifying areas for improvement and guiding policy and operational decisions.

Consistency conditions for bank efficiency analysis in Ghana: A comparison of parametric and non-parametric techniques

This paper extends the concept of methodological crosschecking by examining whether bank efficiencies computed by the two frontier techniques, stochastic frontier analysis (SFA) and data envelopment analysis (DEA), are consistent. The study used a panel of 220 unbalanced observations from 27 Ghanaian banks between 2007 and 2016 to estimate cost and technical efficiencies and check for consistency using five criteria: efficiency distribution, ranking, ability to identify best or worst banks, stability of efficiencies, and relationship with accounting ratios. The results suggest that there is no consistency in the way parametric and non-parametric techniques rank or identify the best or worst banks. Also, there exists a weak relationship between the efficiency scores generated by both SFA and DEA and the non-frontier accounting ratios of Ghanaian banks. This suggests that the latter may contain some exogenous variables that make them weak measures of efficiency and should be used with caution, especially for bank supervision. However, the SFA approach yielded efficiency scores that were comparatively more stable over time. Therefore, the study concludes that the SFA approach is more practical and thus more appealing for regulatory purposes in Ghana due to the relatively consistent efficiency scores under the SFA approach compared to those under the DEA.

Stage stochastic incremental data envelopment analysis models and applications

Data envelopment analysis (DEA) is a mathematical programming method that can evaluate the relative efficiency of multiple inputs and multiple outputs of a decision-making unit (DMU). The classical DEA model assumes that inputs and outputs are determined. However, there are some applications where the inputs–outputs are stochastic. In practice, it is important to evaluate stage performance. It is essential to eliminate the effect of preceding stage inputs (outputs) on stage performance in order to accurately assess stage performance. In this paper, we propose stage stochastic incremental DEA models that integrate two different kinds of inputs and outputs. The first kind of model takes into account the assessment of stage efficiency when determinate incremental inputs and stochastic incremental outputs are applied at the beginning and end of the stage. The second kind of model uses stochastic incremental inputs–outputs to evaluate stage efficiency. To verify the efficacy of the suggested models, the first kind of model is applied to assess the stage financing efficiency of 15 energy-saving and environmental protection clean enterprises (ESEPCEs). The second kind of model is applied in assessing the stage investment efficiency of 15 ESEPCEs. The empirical results show that the proposed models not only eliminate the effect of prior performance but also more accurately assess stage efficiency in a stochastic environment.

Chance constrained directional models in stochastic data envelopment analysis

We construct a new family of chance constrained directional models in stochastic data envelopment analysis, generalizing the deterministic directional models and the chance constrained radial models. We prove that chance constrained directional models define the same concept of stochastic efficiency as the one given by chance constrained radial models and, as a particular case, we obtain a stochastic version of the generalized Farrell measure. Finally, we give some examples of application of chance constrained directional models with stochastic and deterministic directions, showing that inefficiency scores obtained with stochastic directions are less or equal than those obtained considering deterministic directions whose values are the means of the stochastic ones.

A hybrid stochastic data envelopment analysis and decision tree for performance prediction in retail industry

Assessing the retail industry's efficiency is pivotal for economic growth and corporate productivity. This study employs a novel approach, utilizing a regression-based Stochastic Data Envelopment Analysis (SDEA) model, Balanced Scorecard (BSC), and Decision Tree. The integration of these methods is a pioneering effort in the retail sector. This is a data-driven decision-making framework, aiding managers in predicting efficient and inefficient Decision-Making Units (DMUs). Results from a case study in 44 retail store chains in Iran indicate that the accuracy of the SDEA model is 99%. The Decision Tree highlights low branch efficiency due to a low customer count, a unique finding in comparison to prior studies.

Lending business models and FinTechs efficiency

The aim of the study is to analyse which managerial issues can be considered the main efficiency drivers for all Italian FinTechs engaged in lending. We measure their efficiency in the period 2020–2022 via Stochastic Data Envelopment Analysis. The main determinants seem to be ROA and cost-to-income ratio; this means that the ability to control both the business risk level and costs is crucial for FinTechs’ managers and other players interested in M&A deals in this industry. The results are useful for FinTechs, other financial players, regulators and supervisors in defining homogeneous rules in the lending sector.

Environmental performance evaluation in the forest sector: An extended stochastic data envelopment analysis approach

This study addresses the global concern about undesirable outputs in the Forest Sector. We propose two innovative models, namely a directional weak disposable DEA model and an extended stochastic DEA model, to measure environmental efficiency. These models make a significant contribution to the field by specifically assessing the uncertain environmental efficiency of the forest sector. We validated our proposed models by conducting an empirical application using the United Nations Economic Commission for Europe (UNECE) forest sector dataset. The study examines important outputs such as above ground biomass stock, export unit prices of industrial roundwood, wood removals (desirable outputs), and CO2 emissions from wildfires (undesirable output). The results demonstrate that our novel stochastic weak disposability DEA model outperforms traditional approaches when the second scenario is applied. Specifically, the average technical efficiency (TE) score decreases to 0.92, and the number of efficient units reduces to 27, representing an approximate improvement of 55 %. Furthermore, the reduction rate of CO2 emissions is 4.09 % lower than the benchmark. Hence, our extended novel stochastic weak disposability DEA approach enhances the assessment of efficiency and inefficiency in decision-making units, contributing to the mitigation of risk and uncertainty. It also improves overall environmental performance in forest management.

A firm-specific Malmquist productivity index model for stochastic data envelopment analysis: an application to commercial banks

In the data envelopment analysis (DEA) literature, productivity change captured by the Malmquist productivity index, especially in terms of a deterministic environment and stochastic variability in inputs and outputs, has been somewhat ignored. Therefore, this study developed a firm-specific, DEA-based Malmquist index model to examine the efficiency and productivity change of banks in a stochastic environment. First, in order to estimate bank-specific efficiency, we employed a two-stage double bootstrap DEA procedure. Specifically, in the first stage, the technical efficiency scores of banks were calculated by the classic DEA model, while in the second stage, the double bootstrap DEA model was applied to determine the effect of the contextual variables on bank efficiency. Second, we applied a two-stage procedure for measuring productivity change in which the first stage included the estimation of stochastic technical efficiency and the second stage included the regression of the estimated efficiency scores on a set of explanatory variables that influence relative performance. Finally, an empirical investigation of the Iranian banking sector, consisting of 120 bank-year observations of 15 banks from 2014 to 2021, was performed to measure their efficiency and productivity change. Based on the findings, the explanatory variables (i.e., the nonperforming loan ratio and the number of branches) indicated an inverse relationship with stochastic technical efficiency and productivity change. The implication of the findings is that, in order to improve the efficiency and productivity of banks, it is important to optimize these factors.

Changes in efficiency and physical size of container ports: An integration of genetic matching and stochastic data envelopment analysis

Benchmarking container ports of different physical sizes and figuring out the relationship between size and efficiency is complex due to the heterogeneous environment. In dealing with heterogeneity, the selection bias is often overlooked. Therefore, this study proposes an integrated multivariate genetic matching and stochastic DEA algorithm to evaluate the efficiency of container ports. It paves the way for container ports that differ in a selected feature, such as size, to be benchmarked using well-balanced clusters. Thus, the port managers can identify the most similar-featured peers in benchmarking with DEA or alternative models to acquire robust estimates without dependence on a longitudinal data set. The results of the model applied to international container ports imply the increase in size impacts efficiency negatively, while connectivity does positively, which contradicts the commonly held perception of stakeholders that the larger the container ports, the more efficient. That is, well-managed small container ports can also be as efficient. Therefore, it is concluded that the results of integrating genetic matching into the performance measurement provide beneficial inferences. In future research, clustering the observed container ports according to a specific feature and balancing clusters with multivariate genetic matching can provide valuable insights into the industry.

Efficiency assessment of public sector management and culture-led urban regeneration using the enhanced Russell-based directional distance function with stochastic data

Generic data envelopment analysis (DEA) models are based on deterministic input and output. However, input and output vectors are often interrupted by random factors, such as measurement errors and data noise, in real economic situations. This study proposes a new chance-constrained network DEA model based on the modified directional distance function (DDF) and enhanced Russell measure (ERM) model for assessing government management and culture-led urban regeneration. In addition to exploring the randomness of data, this study integrates the advantages of both ERM and DDF and considers the inefficiency level from a non-oriented viewpoint, the direction vector, and each input and output simultaneously. Each input and output of the two production stages can use non-radials to measure efficiency. Results show that the urban-rural gap has gradually widened since 2015. To validate the legitimacy of the model, this study utilizes the bootstrapping method to verify the results of the stochastic network DEA model and the conventional two-stage network DEA approach. This study also considers different alpha values as basis for comparison to confirm whether the results obtained differ by uncertainty level.

Lending activity efficiency. A comparison between fintech firms and the banking sector

The FinTech phenomenon is undoubtedly increasingly changing the morphology of the global financial system, as well as the existing competitive levers in particular sectors, including lending. The aim of this study is to offer a comparative analysis of the level of efficiency exhibited by FinTech firms operating in this sector with that of banks, which have traditionally carried out this activity. We measure efficiency levels by implementing the Stochastic Data Envelopment Analysis (SDEA). The study, referred to 2021, analyses a data set composed of all the Italian FinTech firms engaged in the lending business and all the Italian banks. We find higher efficiency levels for banks compared to FinTech firms. The results are certainly interesting both at corporate level and for regulatory purposes.

Integrated stochastic multicriteria acceptability analysis and data envelopment analysis with fixed-sum outputs: an application for evaluating participating nations in the Winter Olympics Games

Along with the popularity of Olympic Games, evaluating and ranking participating nations has become one of the hottest issues in recent years. Data envelopment analysis (DEA) is extensively used for participating nations’ performance evaluation, while the prior literature mainly addresses the maximum efficiency by ignoring the weights diversity, which might lead to unreasonable and biased results. By considering the fixed-sum medals constraint, this paper proposes an integrated approach of stochastic multicriteria acceptability analysis (SMAA) and DEA for evaluating the participating nations. The proposed approach addresses the weights diversity and multiple equilibrium efficient frontiers by investigating the overall feasible space, and two strategies are considered to guarantee the fairness concern for calculating the efficiencies. Besides, the proposed approach provides both rank acceptability and holistic acceptability index to obtain the complete ranking from the best to the worst and to get the relative ranking relationship in the overall feasible space. The proposed approach can provide more reasonable performance measurements since it explores different rank positions and improves the whole satisfaction of all participating nations. Finally, the proposed approach is applied to empirically analyze the participating nations in the Winter Olympic Games from 2010 to 2022 and some valuable analytical results are provided.

A stochastic approach for evaluating production planning efficiency under uncertainty

<span lang="EN-US">Planning production is an essential component of the decision-making process, which has a direct bearing on the effectiveness of production systems. This study’s objective is to investigate the efficiency performance of decision-making units (DMU) in relation to production planning issues. However, the production system in a manufacturing environment is frequently subject to uncertain situations, such as demand and labor, and this can have an effect not only on production but also on profit. The robust stochastic data envelopment analysis model was proposed in this study with maximizing the number of outputs as the objective function thus means of handling uncertainty in input and output in production planning problems. This model, which is based on stochastic data envelopment analysis and a method of robust optimization, was proposed with the intention of providing an efficient plan of production for each DMU of stage production. The model is applied to small and medium-sized businesses (SMEs), with inputs consisting of the cost of labor, the number of customers, and the quantity of raw materials, and the output consisting of profit and revenue. It has been demonstrated through implementation that the proposed model is both efficient and effective.</span>

Marginal rates of technical changes and impact in stochastic data envelopment analysis: An application in power industry

Marginal rates of technical changes and marginal impact are useful tools to calculate the trade-offs between production factors in a production process. In the data envelopment analysis (DEA) framework, the calculation of these trade-offs (marginal rates of technical substitution, marginal rates of transformation, marginal productivity and marginal costs) using the existing deterministic approaches may be sensitive to uncertainty and variability of the input and output data. Therefore, in this contribution, we introduce a stochastic DEA model based on chance-constrained programming to develop a measure of the marginal rates of firms facing data uncertainty. In this contribution, chance-constrained programming is used to develop a procedure to calculate these trade-offs. Our proposed stochastic procedure is applied to sample data on 31 power plants. The empirical results on marginal rates obtained from our proposed stochastic programs revealed that the results are different at various tolerance levels of chance constraints.

Estimation of Technical Efficiency in Tea Farms at Plantation Level: A Review

The systematic literature review of 111 abstracts has been conducted to comprehensively compile the empirical studies of 21 complete text papers from all over the globe in context to estimation or determination of the technical efficiency (TE) at plantation level of tea production system (TPS), by adopting two methodologies viz., stochastic frontier analysis (SFA) and data envelopment analysis (DEA) during the period 2012-2022. Investigation from these empirical studies revealed that the average TE (TEmean) of tea growers (TG)s all around the globe computed by using both the approaches is around 67.98%, which showcased that the TGs have ability to increase the green tea leaf (GTL) production by 32.02% through better utilization of available resources and technology. The influence of various factors on TE of these TGs had contradictory outcomes, which broke new ground for future research. Computation of TE will enable an investigator to benchmark the best performing TGs in a particular area, which may be referred to by the inefficient TGs to enhance their performance.

Estimating the potato farming efficiency: A comparative study between stochastic frontier analysis and data envelopment analysis.

The government of Bangladesh has been trying to encourage potato consumption to reduce pressure on rice consumption and earn foreign currency along with ensuring zero hunger that helps to achieve the Sustainable Development Goal. It is necessary to use farmers' resources and current technology more efficiently to meet the demand. Therefore, the authors aimed to evaluate the farm-level efficiency of potato farming in Bangladesh. The Cobb-Douglas Stochastic Frontier Analysis (SFA) and the input-oriented Data Envelopment Analysis (DEA) methods are used to compute farm-level technical, allocative, and economic efficiencies and inefficiency of potato farming. The primary data were collected through interviews of 300 potato farmers from Munshigonj, Rangpur, Dinajpur, and Joypurhat districts of Bangladesh. The findings revealed that the efficiency score of the SFA model is higher than the DEA model, which implies that the SFA frontier fits better than the DEA frontier. In the case of DEA, variable returns to scale (VRS) technical efficiency (TE) enveloped data more closely than constant returns to scale (CRS) TE. Results of efficiency suggest significant economic, technical, and allocative inefficiencies in potato farming and there is a scope to increase potato production levels through efficiency improvement. Inefficiency analysis shows that infrastructure and socio-economic factors jointly influence potato production variability. The authors suggest for using the SFA to find efficiencies in the agriculture sector. To achieve efficiency in potato production, the government needs to pay attention for improving the allocative and economic efficiencies along with emphasizing to choose the appropriate technology and efficient use of resources for the scale of operation.

Evaluation of dynamic eco-efficiency of water companies: the influence of non-revenue water and water supply interruptions

Benchmarking the dynamic performance of water companies is a powerful tool to develop sound policies. However, it requires the use of reliable and robust methods and quality of data. In this study, the stochastic nonparametric envelopment of data (StoNED) method was employed to assess productivity change and eco-productivity change of a sample of Chilean water companies over the period 2010–2018. The StoNED approach overcomes the main limitations of stochastic frontier analysis and data envelopment analysis techniques which are the most commonly used to evaluate performance of water companies. Productivity change metric focused on the economic performance, whereas eco-productivity change evaluated economic and quality of service performance by including non-revenue water and unplanned water supply interruptions. Hence, this study analyzed the impact of quality of service variables on the dynamic eco-efficiency of water companies. The annual average productivity change and eco-productivity change estimations were 0.18% and −0.14%, respectively. It was evidenced that on average the Chilean water companies have put as priority to reduce operational costs than to improve quality of service. In both performance metrics, efficiency change was the main contributor to performance changes, whereas the role of technical change was negligible. It was illustrated that full private water companies performed better than concessionary companies and the public company. Reducing non-revenue water and unplanned water supply interruptions was particularly costly for concessionary water companies. The methodology and conclusions of this study are of great interest for water regulators and water utilities to improve the benchmarking process in the provision of drinking water services.

A novel stochastic semi-parametric frontier-based three-stage DEA window model to evaluate China's industrial green economic efficiency

Traditional three-stage data envelopment analysis (DEA) models do not consider the problem of functional form and multicollinearity. This study develops a new stochastic semi-parametric frontier-based three-stage DEA model. The frontier incorporates the effects of both external environmental factors and statistical noise on efficiency. We adopt the StoNED (stochastic non-smooth envelopment of data) approach and use the quasi-likelihood estimation method to estimate the parameters of inefficiency term and stochastic noise. We conduct Monte Carlo experiments to examine the performance of the new frontier under different circumstances. Our results show that the new frontier provides a more realistic and accuracy estimator for efficiency measures. An empirical analysis is used to evaluate green economic efficiency (GEE) in China. We empirically compare different models and the results show that external environmental factors cause significant differences. We provide each provincial average GEE evaluated by the improved QLE-StoNED model, which are outperforms compared with other recently developed estimators. And a gradient difference emerges in the GEE among the eastern, central and western areas of China. The results also offer practical implications for the harmonious development of industrial production and a green economy in China.