Sessions & Abstracts


Opening Session 1

Zilla Sinuany-Stern, Ben Gurion University of the Negev

Opening Session 2

Menachem Ben-Sasson, President, Hebrew University of Jerusalem

Yishay Yafeh, Dean of the Business School, Hebrew University of Jerusalem

Boaz Golany, Vice President, Technion

Yohanan Danino, Police Commissioner

S1: Electricity 1

Rajiv Banker, Temple University: On production frontiers and incentive regulation in the electric power industry.

S2: Electricity 2

Endre Bjørndal, Mette Bjørndal and Thore Johnsen, NHH Norwegian School of Economics, Norway: Calibrating Cost Norms for Electricity Networks based on DEA Efficiency Analyses

A principal objective of the regulation is to give incentives for cost efficiency in operations and investments, including to facilitate the best possible organization of individual companies and the industry in total. We consider how to establish cost norms and revenue caps for electricity companies based on DEA efficiency analyses. We discuss whether and how the super-efficiency concept can be used to improve the incentive properties of the benchmarking model. Also, we discuss whether calibration of the total industry cost norm is necessary in order to ensure that companies with average efficiency earn the regulatory rate of return. We also discuss how this calibration could be done, and whether the choice of method has any consequences for, e.g., the time profile of the cash flow of the companies.  We use data for Norwegian distribution and regional transmission companies in our analyses.

Rajiv Banker, Ana Lopes and Shunlan Fang: Drivers of Distribution Costs in a Brazilian Electric Power Company.

This study develops and estimates an econometric model relating the costs of distribution of electric power to the drivers of those costs in order to understand reasons for differences in costs and apparent inefficiencies of distributor companies.  Costs are analyzed into various maintenance and customer service costs.  Drivers of those costs include variables reflecting demographic and economic differences, topology of the area of operation, the quality of the services provided and the demands of society.  This model is designed for general application to all Brazilian energy distribution companies.  Estimation of the detailed cost model capturing the cause-effect relationships rather than the coarse and unjustified model employed by the Brazilian regulator provides a theoretically correct basis for benchmarking the Brazilian electricity distributors and determining the existence and sources of inefficiencies.  Until the Brazilian regulator starts collecting the requisite data from Brazilian electricity distributors, we estimate our initial detailed cost model using data collected from the largest Brazilian integrated energy company – Centrais Elétricas de Minas Gerais (CEMIG) — separately for the 14 regions it operates in the province of Minas Gerais.


S3: Banking

Mette Asmild, University of Copenhagen, Denmark: Regulating bank risk using weight-restricted DEA.

The recent financial recent crisis demonstrated how banks’ risks are associated with the mixes within their funding and/or asset portfolios. In order to appropriately measure bank performance, efficiency measures that capture this risk element are needed. However, in standard Data Envelopment Analysis models, banks with extreme mixes of inputs (such as whole-sale funding relative to retail funding) or outputs (such as property loans relative to other loans) may be deemed efficient “by default”. Thereby, very risky banks can appear fully efficient as a consequence of their extreme mixes whilst ignoring the associated risks. In a representative sample of over 70 of the largest banks in Europe for the time period 2006-2009, we find an overrepresentation of banks that required government bail-out amongst the banks with relatively extreme mixes of inputs and/or outputs, i.e. the ones potentially efficient by default. To address this problem, we propose a weight-restricted DEA model where the weights attached to inputs and outputs by all banks are restricted to the ranges used by a set of model banks. We then investigate how such weight restricted DEA models can be used as a regulatory tool which, unlike the various BASEL accords, simultaneously consider the various inputs and outputs, as well as explicitly control for the risk originating from the input- and output-mixes.

Angela Tran Kingyens and Joseph C. Paradi, University of Toronto: Regulatory Issues in the Canadian Financial Services Industry.

Canadian Financial Services regulators enforce the “Know Your Client” rule upon Canadian financial institutions and their advisors.  The fundamental reason is to prevent inappropriate investments for people whose risk taking ability and temperament do not allow certain investments.  On the other hand, there should be room for people who are both willing and capable to take risks to be able to do so. This work was designed to address the need for Financial Institutions to be able to determine the risk levels their customers have.  The problem is most acute in cases where the investor does his/her investing on the Internet.  The work was accomplished by creating and using a DEA friendly questionnaire which then gave a risk propensity score to the institution.

S4: Public Transport

Bart Roets and Johan Christiaens, Infrabel/University of Ghent: Efficiency benchmarking of railway traffic management.

Railway infrastructure managers are urged by European railway directives and national austerity measures to increase their efficiency levels. Previous academic research on the subject of railway infrastructure efficiency focused on “asset management”, and was mainly based on parametric techniques. Our study, supported by Belgium’s railway infrastructure manager Infrabel, fills a gap in the present literature by examining the efficiency of railway “traffic management”.
Using archival data from accounting and operations databases, we develop a DEA model which assesses the relative efficiency of Infrabel’s electronic signal boxes. Possible sources of inefficiency, such as network complexity and traffic density, are also examined.


Álvaro Costa, Teresa Stanislau, Sebastian Ebert, University of Porto: Management Decisions in Public Transport: Impacts on Efficiency and Effectiveness – a DEA-based case study on STCP, Porto.

The adequate functioning of public transport (PT) highly contributes to the quality of life in cities. However, current trends, such as more people living on less space, increasing energy costs, financial shortages or regulatory rules require decision makers more than ever to act economically sound. Thus, managerial practice, decisions taken and strategies applied in the operational framework are of high importance for evaluation. The set of management choices in public transport refers to network-, finance-, personnel-, fleet-, schedule and service-related decisions. However, in terms of performance, decision makers often have to face uncertainty about the true outcome of their strategic choices and operational judgements. Thus, the relation between managerial decisions in PT and the provider´s performance in terms of efficiency and effectiveness becomes an essential area of research. The objective of this paper is to present a case study from Porto-based (Portugal) operator Sociedade de Transportes Colectivos do Porto (STCP) on how to evaluate the relation between managerial decisions and economic indicators. By means of Data Envelopment Analysis (DEA) the company’s performance results between 1946 and 2011 are analysed and the impact of management acts as well as the consequences and conclusions implied are discussed. The approach aims at identifying the magnitude of decisions and their time-lags within the operational PT framework. By doing so, an impetus for wider research will be set. Further results might be translated into policy recommendations, e.g. to design regulatory frameworks which foster PT systems to perform economically and socially at their best.

S5: Infrastructure

Konstantinos Triantis, Virginia Tech University: The Use of Network DEA for Transportation Infrastructure Systems.

We propose an evaluation approach for a novel travel demand management strategy known as the downtown space reservation system (DSRS). This approach takes into account three perspectives, i.e., transportation service provider’s, the user’s, and the community’s and is based on network-Data Envelopment Analysis (DEA) where the perspectives are inter-related through intermediate inputs/outputs. Two types of network-DEA models (radial and slacks-based models) are considered. An example is provided using data propagated from a microscopic traffic simulation model of the DSRS. The results show that individual node performance can drive network DEA performance and that this information can inform future designs of the DSRS.

Arik Sadeh, HIT, Holon, Israel: Efficiency and Success Measurement of R&D Projects

The concept of Data Envelopment Analysis (DEA) is used to compare the relative and absolute performance of R&D projects. These projects are difficult to evaluate by the relationship between their inputs and outputs and are usually not evaluated by their economic success. Success of R&D projects is measured by a set of several output variables. Multiple inputs are considered, for measurement of efficiency and success of such projects. This analysis is performed with respect to associated inputs and predetermined conditions such as the level of novelty, complexity, technological uncertainty, and pace of the projects.  Over 100 R&D projects are evaluated and analysis of their success and efficiency is conducted.

M1: Theory

Finn Forsund, University of Oslo, On the effectiveness of the public sector.

The distinction between the concepts outputs and outcomes can be made operational based on the degree of control a public service producer has over its production activity. Resources are transformed into service outputs under the control of the service provider while outcomes represent some higher social goals than outputs and are determined by the outputs and exogenous variables. Production of outcomes is outside the control of the organisation. Transformation of resources into service outputs are modelled traditionally way, but the transformation of service outputs to outcomes is modelled by factorially determined multioutput production functions. Outcomes range from private goods produced by the individual consumers as in a household production function approach, to pure public goods. The link to efficiency
measurement is provided by introducing the concept of a benchmark frontier technology for the type of production in question. A new measure of overall preference efficiency is introduced and its decomposition into output-oriented technical efficiency and output mix efficiency is shown. The rather monumental task of providing the necessary information for calculating mix efficiency is

H. David Sherman and Joe Zhu, Northeastern University: Analyzing Performance in Service Organizations.

Just as baseball teams have been increasingly relying on rigorous quantitative analyses, so have many businesses. In particular, a growing number of service organizations have been investigating the use of a sophisticated linear programming technique called DEA, or Data Envelopment Analysis. (In this article, we use the term “balanced benchmarking” to denote DEA.) The technique enables companies to benchmark and locate best practices that are not visible through other commonly used management methodologies. Today, balanced benchmarking can be utilized by anyone with Microsoft Excel, but it was not always so easy. When it was first introduced in the 1980’s (Charnes, Cooper & Rhodes), balanced benchmarking was an academic tool for measuring and managing relative efficiency of peer organizations. Balanced benchmarking simultaneously considers the multiple resources used to generate multiple services, along with the quality of the services provided.  For example, bank branches can use six or more types of resources and provide 20 or more types of services, all of which are considered with balanced benchmarking. By combining this information, balanced benchmarking provides unique insights about best practices and opportunities to improve productivity and profitability, information not available with other techniques. Balanced benchmarking provides managers with a sophisticated mechanism to assess the performance of different service providers—comparing, for example, the London and Tokyo offices of a global advertising agency—by going well beyond the crude metrics and ratios such as profitability and account billings per employee. From the results of balanced benchmarking, a company can identify its least efficient offices or business units, and it can assess the magnitude of the inefficiency and investigate potential paths for improvement that the analysis has identified. Moreover, executives can study the top-performing units, identify the best practices, and transfer that valuable knowledge throughout the organization to enhance performance. Lastly, balanced benchmarking enables companies to test their assumptions, particularly before implementing cost-cutting initiatives that might inadvertently be counter-productive.

Thijs ten Raa, Tilberg University: Performance measurement in an input-output framework.
This paper fruitfully combines two complementary theories: DEA and input-output analysis. Our point of departure is the theory of the consumer, who maximizes utility subject to a budget constraint. His well-being can be measured by the change in the consumption bundle, valued at constant prices. Input-output analysis and DEA will be invoked to impute the change in this bundle to technical change, a terms-of-trade effect and two types of efficiency change. The analysis is extended to environmental economics.

M2: Public Services

Ekaterina Yazhemsky, David Grader-Sageev and Mali Sher, Hebrew University: Setting Performance Targets for Police Units

First time in the academic literature, we describe the operative application of DEA in the process of performance improving of police stations. In this study we present the methodology used to determine the required annual achievement of each investigation unit to reach organization goals. These targets have been incorporated in the Performance Measurement System of Israel Police called “Mifne” in the period 2012-2013. The developed approach is based on a combination of multidimensional clustering, Data Envelopment Analysis, simulation and Balanced Scorecard models. Results of the proposed approach provide meaningful benchmarks with comparable peer units and target values that are potentially achievable in the annual terms for both efficient and relatively inefficient units, steering the organization in the direction of the top priority goals.

Zilla Sinuany-Stern, Lea Friedman, and Dov Chernechovsky, Ben Gurion University, Beer-Sheva, Israel: Macro and micro analyses of hospitals efficiency

Data Envelopment Analysis (DEA) and its variants are widely used for measuring hospitals efficiency. In macro analysis of hospitals’ efficiency the Decision Making Unit (DMU) is the hospital as a whole. In micro analysis the DMU may be a ward within a hospital or even less than that. Sometimes the macro and the micro levels share the same type of inputs and outputs. In those cases it is interesting to verify if there is consistency in the efficiencies of the macro vs. the micro levels. However, due to lack of data on the macro level, often the micro level has more detailed data. It is interesting to verify the consistency of the efficiencies of the macro versus the micro level. We show via several examples that these consistencies are questionable in both cases. On the macro level, we have 21 hospitals the inputs are: the number of standard beds, while the outputs are: the number of hospitalization days and number of discharges. In the first micro example for each of the main 5 wards we measure the efficiency across the 21 hospitals using the same inputs and outputs as in the macro model. In some wards the results were consistent with the macro level, while in some other cases, they were not consistent. In the second example, within a given hospital we looked at the surgery unit which serves several wards and we measured the efficiency of some 24 departments which use the operation rooms. In this case the inputs and outputs were more detailed than in the previous examples. Evidently, again the efficiencies were not always consistent with the second example.


Amos Bick, Leonid Gillerman, Yossi Manor and Gideon Oron,  Ben Gurion University, Beer-Sheva, Israel: Implementation Data Envelopment Analysis for Optimal Operation of Ultrafiltration Systems for Effluent Reclamation.

One of the major problems in operating a combined UltraFiltration (UF) and Reveres-Osmosis (RO) membrane systems is identifying optimal working regime for the UF stage. It is associated with adequate decision-making regarding the flow rates, back flushing strategy and the use of chemical reagents to minimize scaling and bio fouling. The appropriate decision-making can be maintained upon analyzing experimental field data. It includes the permeate flow rates, permeate turbidity, and pressure head loss across the membranes. There is a need for a common definition and measuring of assessing the information quality and data-effectiveness concerning the UF performance. Data quality is conceived as a quantity of aggregated values of multiple subjective and objective criteria.  Fundamentally, efficiency of a system performance for various production scenarios can be defined as the ratio between outputs and inputs. Therefore it is imperative to consider multiple inputs and outputs, leading to optimal system performance. Data Envelopment Analysis (DEA) can be implemented for better assessment of the UF system performance, leading ultimately to optimal working conditions. DEA differs from other comparative normalization methods that it does not focus on the complete set of data, but rather on individual Decision-Making Units (DMU). Accordingly, a discrete piecewise frontier of efficient DMUs’ is developed, entitled as an “Efficient Envelope”. In view of that, a production frontier expresses technological relationships that describe optimal output range that the membrane can generate. Implementing the DEA allows to determine information quality of input data (e.g. trans-membrane pressure, feed temperature and feed turbidity) and output data (e.g. plant recovery, permeate flux and turbidity removal). Data normalization of the field results implementing the DEA with reference to the technical efficiency is therefore essential. Field results from an integrative pilot plant system and an agricultural field demonstrates the potential of implementing the DEA for optimal operation of the UF system.

T1: Policy

Marina Toger, Dan Malkinson, Izthak Benenson, Danny Czamanski, Technion: Regulation of urban nature and urban spatial dynamics

The dynamics of urban nature is intimately linked to the dynamics of urban built environment. The paper traces and studies the evolution of two interlinked systems. The evolution of the built environment is modeled as a self-organizing system operating within the context of extant regulatory regimes. Changes in regulatory actions are examined in terms of impacts on urban morphology. Changes in urban open spaces in Haifa are examined in terms of size, form, and relative position to each other, distance from the city edge and as a network. Among other things the paper presents analyses of network connectivity and clustering. Extent of connectivity influences how well the system is functioning, as it facilitates organism movement across the urban landscape. Growing urbanization increases fragmentation and loss of biodiversity. As undisturbed areas become scarcer, open spaces and the surrounding matrix need to provide the habitat requirements of various species. The paper presents initial data concerning presence of wildlife collected along two urban-rural gradients in Haifa. Preliminary analyses check these findings against the network morphology of the open spaces.

Nir Sharaby, Yoram Shiftan, Technion: The impact of fare integration on travel behavior and transit ridership.

Integrated, high quality and accessible transit systems are essential for attracting travelers to shift from private to public. This change in personal behavior is desired in order to reduce congestion and air pollutant emissions in city centers. This paper focuses on evaluating the impact of fare integration on transit ridership and travel behavior, using the city of Haifa, Israel, as a case study. The city’s new, integrated fare policy changed the historically complex per-boarding system to a simple five-zone fare system with free transfers, reducing fares for many passengers. Using fare-box data, on-board surveys and travel-behavior model estimation, we show that the new fare policy managed to negate the downward trend in transit ridership. Fare-box data showed a significant increase in single ticket sales of up to 25% over the first year following the launch of the reform; the survey’s results pointed to an increase of 7.7% in passenger trips and 18.6% in boarding. The number of passenger boarding per trip increased from 1.38 to 1.52, implying that people were utilizing the free transfer option and enjoying a wider range of routes. The model results showed that fare reduction was a significant factor in attracting transit users, and that the public transport reform had three important contributions: first, it encouraged travelers to shift from private cars or taxi to buses; second, it created new trips, offering more opportunities for activity participation; and third, it increased travel options by allowing travelers to choose a better route.

Joel Zhengyi Shon, Tainan University of Technology: The Effective Tools to Evaluate the Productivity of Public Owned Transportation Business.

The mission of transportation business is far more than profit maximization when economic externalities are taken into consideration. Regulators usually need to manage some business activities to make sure public interests are secured. In most cases, regulators also need to compensate the operators for their losses of providing transportation services. Since money making can hardly be the sole operation goal of transportation business, investors from private sectors are usually reluctant to run the businesses of public transport because of poor resources allocations. As a result, there are a lot of public owned companies in the highly twisted business environment in the transportation sector. It’s not surprising there are a great number of public owned airlines, airports, maritime liners, sea ports, buses, subways, and railways throughout the world. These companies, owned by the government or consortium of public interests to provide transportation service while carrying policy purposes in the same time, are very difficult to be as productive as private sector when the productivity was measured by financial indicators. This research aims to find out some proper indexes to evaluate the productivity of public owned transportation business. Airports and sea ports are two examples to be analyzed. After the suggested evaluation tools are introduced, some thoughts and ideas will be discussed for further studies.

T2: Aviation 1

Yeon Myung Kim, OECD

Moshe Givoni, Tel-Aviv University: Airline and railway (dis)integration.

Mode substitution – from air transport to rail transport is a popular idea promising to be a win-win policy in the context of sustainable transport, airport congestion and revitalized rail industry. When the model of Airline and Railway Integration was suggested it promised to offer important benefits. The research on airline and railway integration ended with the assumption that air and rail integration will be widespread as its benefits are obvious and its potential will only increase. Close to ten years on, when the HST network around the world is fast developing and expanding and the notion of mode substitution is still apparent in contemporary transport policy, for example that of the EU, air and rail transport are still largely disintegrated. This is a missed opportunity. A missed opportunity to advance sustainable transport, a missed opportunity for a different approach to air transport provision, and even more so a missed opportunity for advancing the renaissance of the railways. Although not a panacea for any of the above challenges, and although it could be counterproductive for sustainable transport, the idea of airline and railway integration holds great potential. Trying to understand why this opportunity is being missed is the main aim of this presentation. The conclusion is that the dominance of ‘competition’ rather than complementarity in the regulatory debate is the main barrier to airline and railway integration, and likewise integration between other parts of the currently fragmented transport network.

Kenneth Button, George Mason University: What do we really know about the efficiency of the European air traffic control system.

The paper examines the relative efficiencies of alternative institutional structures under which air navigation services are provided. In makes use of panel data from Europe to consider the effects of differing regulatory and financing systems on the provision of these services by deploying both data envelopment analysis and, for reinforcement, stochastic Gaussian methodologies. The evidence suggests that in the short run, the systems that have been commercialized are performing no better than those that have not, although the gap narrows after allowing for spatial autocorrelation.

T3: Aviation 2

Nicole Adler, Hebrew University and Nicola Volta, University of Southern Denmark: DEED, a directional economic-environmental distance function approach

We propose a new data envelopment analysis model in order to compute efficiency from an eco-environmental point of view. The directional, eco-efficient distance (DEED) function includes undesirable as well as standard variables hence is suitable for multiple industrial applications. Moreover, the model designs a single frontier which permits a constrained increase in inputs where relevant. In addition, given the prices of inputs, the objective function estimates the maximum monetary savings potential for inefficient decision making units. We show that the DEED function improves on existing models in the literature by substantially reducing negative externalities within a budget constraint and utilizing only efficient decision making units as benchmarks. In an application analyzing the global airline fleet, we estimate the potential to reduce noise and airborne pollutants emitted by aircraft-engine combinations given the current state of aeronautical technology. The global engine-aircraft market is viewed from the regulatory perspective in order to compare the single environmental and operational efficient frontier to that of the airline carriers, and environmental objectives. The results highlight the inefficiencies of the current airline fleets and that the IPCC values of externalities are a magnitude of TEN too low to encourage changes in the global fleet hence the need for government intervention.

Davide Scotti, University of Bergamo, Martin Dresner, University of Maryland, Gianmaria Martini, University of Bergamo and Chunyan Yu, Embry-Riddle: Incorporating Negative Externalities Into Efficiency Assessments of US Airports

This paper analyses the efficiency of 44 US airports for the period 2005-2009. In addition to the conventional outputs (i.e., passengers, flights and cargo), we consider three undesirable externalities of airport activities: delays, noise and local air pollution. We adopt a directional distance function approach and perform a second stage analysis to investigate potential determinants of efficiency. Our base case results with only the positive outputs show that the greater the average aircraft size serving an airport and the larger the dimensions of the airport, the higher the technical efficiency. However, our results are sensitive to the inclusion of the undesirable outputs.  The implications are that the inclusion of these externalities into the calculation of efficiency may impact managerial decisions.


Hans-Martin Niemeier, Hochschule Bremen University of Applied Sciences: On production frontiers and incentive regulation in the airport industry.

There has been a gradual trend towards incentive regulation of most partially and fully privatised airports since the privatisation of BAA in 1986. In ten countries of the European Union, airports are price capped, and examples of price-capped airports exist elsewhere, notably India. However, none of these price caps is a pure price cap in which the X-factor is set independently of the cost of the regulated airport. Typically, hybrid price caps are used and combined with sometimes complex mechanisms like sliding scales, quality incentives and investment obligations, so that the incentive structures of airports become distorted. In this paper we provide an overview of the changes in the governance structure of airports as a result of privatization. Secondly, we analyse how far the regulatory institutions obey the principles of “good” regulation, such as fairness and transparency. Then we investigate the performance of the new regulation in terms of cost and allocative efficiency, by reviewing benchmarking studies. Finally we assess the impact of incentive regulation on productive efficiency using a benchmarking study and an initial econometric analysis.