DEA Bibliography

1. Adolphson, D. L.; Cornia, G. C.; Walters, L. C. Interfaces 1989, 19( 3), 18-26.
Railroad companies in the United States and tax officials often disagree about the appraisal methodology as well as the appraised value of railroads. A frequent area of disagreement is in the measurement of obsolescence, defined as the reduction in value due to changing economic conditions or design limitations. Current methods of measuring obsolescence in railroads systematically overestimate obsolescence. Data envelopment analysis (DEA) can be used to obtain a better measure of obsolescence. The states of Utah and Wisconsin have adopted the DEA method of measuring obsolescence.
Accounting. Railways. Transportation.

2. Adolphson, D. L.; Cornia, G. C.; Walters, L. C. Proceedings of the Seventeenth Annual Program on the Appraisal of Utilities and Railroad Property for Ad Valorem Taxation; Wichita State University.

3. Afriat, S. Kluwer Academic Publishers: Boston, 1988.

4. Ahn, T.; Arnold, V.; Charnes, A.; Cooper, W. W. 1989; Vol. 5.

5. Ahn, T.; Charnes, A.; Cooper, W. W. Socio-Economic Planning Sciences 1989, 22(6), 253-257.

6. Ahn, T.; Charnes, A.; Cooper, W. W. Socio-Economic Planning Sciences 1988, 22(6), 259-269.

7. Ahn, T.; Charnes, A.; Cooper, W. W. Managerial and Decision Economics 1988, 9(3), 251-253.

8. Al-Faraj, T. N.; Alidi, A. S.; Bu-Bshait, K. A. International Journal of Operations & Production Management 1993 , 13( 9), 45-52.
Details a data envelopment analysis (DEA) study of bank branches, where the bank branches are considered as the decision-making units. The Charnes, Cooper and Rhodes (CCR) ratio is used to evaluate the relative efficiency of each branch. The relative efficiency is computed utilizing specific input and output factors. The study has been applied to a set of branches of one of the largest commercial banks in Saudi Arabia in an attempt to evaluate the relative efficiency of these branches in order to improve the quality of services and utilize the available resources more efficiently.
banking. decision theory. management science.

9. Ali, A. I. Computers, Environment and Urban Systems 1990, 14( 2), 157-65.
The characteristics of the linear programs arising in data envelopment analysis (DEA) necessitate robustness and accuracy of computation. The author discusses issues of accurate and robust computation for DEA focusing on three data envelopment analysis models: CCR, BCC and ADDITIVE.
Administrative data processing. Decision theory. Linear programming.

10. Ali, A. I. European Journal of Operational Research 1993, 64( 1), 61-67.
Discusses issues related to an efficient computational methodology for performing data envelopment analysis focusing on three data envelopment analysis models: CCR, BCC, and ADDITIVE. It introduces constructs that facilitate efficiency of computation in solving a sequence of as many linear programs as there are decision making units. Computational testing with real-world data sets with up to 533 decision making units indicates that employing these constructs reduces computation time significantly.
decision theory. mathematical programming.

11. Ali, A. I.; Cook, W. D.; Seiford, L. M. Management Science 1991, 37( 6), 733-8.
Golany (1988) proposes an interesting extension of data envelopment analysis (DEA) to the situation where ordinal relations among weights corresponding to certain dimensions exist. However, the development of the extended model contains mathematical errors and the proposed equivalence is incorrect. In this paper the authors establish the correct model equivalence and prove that weak ordinal relations require a nonstandard DEA model. It is also shown that the case of strict ordinal relations can be handled with the standard DEA model. These results hold for the CCR, BCC, additive, and multiplicative models and for relations involving both input and output multipliers.
Management science. Mathematical programming. Statistical analysis.

12. Ali, I. A.; Seiford, L. M. Operations Research Letters 1990, 9( 6), 403-5.
Conditions are established under which DEA models are translation invariant. Specifically, an affine displacement does not alter the efficient frontier for models incorporating the convexity constraint. This affords a ready solution to the problems of scaling and the presence of zero values which arise in data envelopment analysis (DEA).
Decision theory. Management science.

13. Andersen, P.; Petersen, N. C. Management Science 1993, 39( 10), 1261-4.
Data envelopment analysis (DEA) evaluates the relative efficiency of decision-making units (DMUs) but does not allow for a ranking of the efficient units themselves. A modified version of DEA based upon comparison of efficient DMUs relative to a reference technology spanned by all other units is developed. The procedure provides a framework for ranking efficient units and facilitates comparison with rankings based on parametric methods.
decision theory. linear programming.

14. Anderson, E.; Y. Chen IIE Transactions 1992, 24(5), 155-165.

15.Arnold, V. L.; Bardhan, I. R.; Cooper, W. W.; Kumbhakar, S. C. Evaluating Public School Performance With Stochastic Frontier Regressions and Data Envelopment Analysis (DEA) University of Texas at Austin: 1993Forthcoming.

16. Athanassopoulos, A. D. European Journal of Operational Research 1996.

17. Athanassopoulos, A. D. European Journal of Operational Research (Submitted) 1995.

18.Athanassopoulos, A. D. Working Paper 1995.
This paper has two objectives: First to review methods regarding the attempts to rank relatively efficient DMUs by DEA. The second is to criticise the so-called super-efficiency models which yield some times dominated solutions. A revised DEA model in provided for assessing superefficiency and comparisons are made using data from a set of restaurants.

19. Athanassopoulos A.D.; Thanassoulis E. Journal of the Operational Research Society 1995, 46(1), UK.
This is a paper concerned with the definition and assessment of the market efficiency of retail organisations. Issues regarding the shop profitability and the long run viability of indivual units are discussed.

20. Banker, R. D. European Journal of Operations Research 1980, 5, 262-266.

21. Banker, R. D. Management Science 1993, 39( 10), 1265-73.
This paper provides a formal statistical basis for the efficiency evaluation techniques of data envelopment analysis (DEA). DEA estimators of the best practice monotone increasing and concave production function are shown to be also maximum likelihood estimators if the deviation of actual output from the efficient output is regarded as a stochastic variable with a monotone decreasing probability density function. While the best practice frontier estimator is biased below the theoretical frontier for a finite sample size, the bias approaches zero for large samples. The DEA estimators exhibit the desirable asymptotic property of consistency, and the asymptotic distribution of the DEA estimators of inefficiency deviations is identical to the true distribution of these deviations. This result is then employed to suggest possible statistical tests of hypotheses based on asymptotic distributions.
decision theory. maximum likelihood estimation.

22. Banker, R. D. Plenum: New York, 1985.

23. Banker, R. D. Harvard University, 1980.

24. Banker, R. D.; Charnes, A.; Cooper, W. W. Management Science 1984, 30(9), 1078-92.

25. Banker, R. D.; Charnes, A.; Cooper, W. W.; Clarke, R. European Journal of Operational Research 1989, 40( 3), 299-308.
Constrained game formulations are provided for DEA (data envelopment analysis) which extend the original (unconstrained) game formulations of R. Banker (1984). Interpretations are also supplied which show how the optimal solutions may be used to improve the efficiency of DMUs (decision making units) being evaluated with repeated uses of DEA. Uses of these ideas in testing for the presence of 'organization slack', are described, as in the literature of organization theory. Possibilities are also suggested for uses of game theory to study tendencies toward specialization by some DMUs, including specializing tendencies which take the form of developing new products or undertaking other activities to escape pressures from the repeated use of relative evaluations that are present in market competition as well as DEA.
Game theory. Management science. Optimisation.

26. Banker, R. D.; Charnes, A.; Cooper, W. W.; Maindiratta, A. Kluwer Academic Publishers: Boston, 1988.

27. Banker, R. D.; Charnes, A.; Cooper, W. W.; Schinnar, A. Management Science 1981, 27(12), 1370-1382.

28. Banker, R. D.; Charnes, A.; Cooper, W. W.; Swarts, J.; Thomas, D. JAI Press: 1989; Vol. 5.

29. Banker, R. D.; Das, S.; Datar, S. M. JAI Press: 1989; Vol. 5.

30. Banker, R. D.; Kemerer, C. F. IEEE Transactions on Software Engineering 1989, 15( 10), 1199-205.
Two opposing views regarding the presence of economies or diseconomies of scale in new software development are reconciled. A production function model of software development that allows for both increasing and decreasing returns to scale is hypothesized, and it is argued that local scale economies or diseconomies depend on the size of projects. Using eight different data sets, including several reported in previous research on the subject, empirical evidence in support of the hypothesis is provided. Through the use of nonparametric data envelopment analysis technique it is shown how to identify the most productive scale size that may vary across an organization.
Project engineering. Software engineering.

31. Banker, R. D.; Maindiratta, A. Econometrica 1988 , 56(6), 1315-1332.

32. Banker, R. D.; Maindratta, A. Management Science 1986, 32(3), 385.

33. Banker, R. D.; Maindratta, A. Management Science 1986, 32(1), 126-135.

34. Banker, R. D.; Morey, R. C. Oper. Res. (USA) 1986 , 34( 4), 513-21.
We evaluate by means of mathematical programming formulations, the relative technical and scale efficiencies of decision making units (DMUs) when some of the inputs or outputs are exogeneously fixed and beyond the discretionary control of DMU managers. This approach further develops the work on efficiency evaluation and on estimation of efficient production frontiers known as data envelopment analysis (DEA). The authors also employ the model to provide efficient input and output targets for DMU managers in a way that specifically accounts for the fixed nature of some of the inputs or outputs. The approach is illustrated using real data, for a network of fast food restaurants.
decision theory. mathematical programming.

35. Banker, R. D.; Morey, R. C. JAI Press: 1989; Vol. 5.
Data Envelopment Analysis (DEA) is useful for evaluating relative efficiency of not-for-profit or governmental organizations because it evaluates technical efficiency of decision making units (DMUs) without requiring a priori knowledge of input or output values. Instead, DEA endogenously generates weights representing a feasible relative value system which provides the highest possible efficiency rating for the DMU being evaluated. These endogenous weights, however, may not be consistent with a realistic ordering of the relative input or output values. On the other hand, allocative efficiency evaluation requres explicit knowledge of relative values. In this paper, we consider an extension of the DEA model to evaluate relative efficiency when information on only the upper and lower bounds of the relative values is available.

36. Banker, R. D.; Morey, R. C. Journal of Operations Management 1993, 11( 1), 81-98.
We consider a fast food chain contemplating acquiring, building, and operating a new outlet at a particular candidate site, with its own particular environment. We offer a normative approach or this type of decision, which arrives endogenously at the best choices for both design and operational features for the site. The final decision as to whether or not to actually acquire, build, and operate the outlet at the candidate site thus depends on several complex connected choices. Our approach integrates nonparametric allocative data envelopment analysis with parametric regression analysis to yield a mathematical programming optimization model. We illustrate the approach with actual data from Hardee's, a fast food chain with over 3100 outlets, and an actual candidate site. The recommendations from the approach are compared with those from the conventional approach. A key advantage of our approach is its contrast to conventional approaches for site selection which ignore the more detailed design and operational choices. The approach identifies the chain's best practices in different environments, thereby enabling management to factor out poor performances due to flawed installation, lack of proper worker incentives, etc. The approach is equally appropriate for other service outlets; e.g., banks, hospitals, or pharmacies.
mathematical programming. operations research. retailing. service industries. statistical analysis.

37. Banker, R. D.; Morey, R. C. Manage. Sci. (USA) 1986 , 32( 12), 1613-27.
Data envelopment analysis has now been extensively applied in a range of empirical settings to identify relative inefficiencies, and provide targets for improvements. It accomplishes this by developing peer groups for each unit being operated. The use of categorical variables is an important extension which can improve the peer group construction process and incorporate 'on-off' characteristics. It relaxes the stringent need for factors to display piecewise constant marginal productivities. In so doing, it substantially strengthens the credibility of the insights obtained. The paper treats the cases when the categorical variable can be controllable or uncontrollable by the manager, for the cases of technical and scale inefficiency of decision making units.
decision theory. management science.

38. Banker, R. D.; Thrall, R. M. European Journal of Operational Research 1992, 62( 1), 74-84.
Generalization of the measure of returns-to-scale from a single number to an interval permits extension of the concept to data envelopment analysis (DEA) data domains with multiple inputs and multiple outputs. The key new approach is a partition of the optimal frontier into three parts corresponding, respectively to increasing, constant, and decreasing returns to scale. These parts are characterized in terms of optimal primal solutions, and optimal dual solutions for both the original Charnes, Cooper, Rhodes model (1978) and the later Banker, Charnes, Cooper model (1984) and relying on concepts developed by R.D. Banker (1984) and R.M. Thrall (1988).
linear programming.

39. Banker, R.; Conrad, R. F.; Strauss, R. P. Management Science 1986, 32(1), 30-44.

40. Banker, R.; Kemerer, C. Proceedings of the 8th International Conference on Information Systems; pp 160-175.

41. Banker, R. D.; Gadh, V. M.; Gorr, W. L. European Journal of Operational Research 1993, 67( 3), 332-43.
Reports the results of an experiment with simulated data that compares the estimation accuracy of two simple and very different production frontier methods: corrected ordinary least squares and data envelopment analysis. The experimental design extends a previously published paper by introducing measurement errors, a factor we show to be critical for comparative analysis of the frontier methods. Both low and high measurement error distributions are used, resulting in 95% error intervals of roughly +or-10% and 40%, respectively, of outputs. Other variations include four inefficiency distributions covering a wide range of behavior; four sample sizes, from 25 to 200, and two piecewise Cobb-Douglas technologies with two inputs and one output each.
estimation theory. least squares approximations. mathematical programming. Monte Carlo methods. production control.

42. Banker, R. D.; Kauffman, R. J.; Morey, R. C. Journal of Management Information Systems 1990, 7(2), 29-54.
This paper presents a new approach to measuring the input productivity gains from information technology (IT) in complex managerial environments. The approach is illustrated in the context of a pilot deployment at Hardee's Inc. of a new cash register point-of-sale and order-coordination technology called Positran. The method employs data envelopment analysis (DEA) and nonparametric production frontier hypothesis testing to determine whether the performance of restaurants that have deployed Positran is better, on average, than for those that have not. The design of the study is of special insterst, because it approximates a controlled experiment. Our results show that Positran helped to reduce the input material costs, since restaurants that deployed the technology were less likely to be inefficient. It is further possible to characterize the class of restaurants for which the relationship holds. Operation efficiency measures such as the ones we have developed provide measures suchs as the ones we have developed provide managers with the opportunity to implement deployment strategies for new ITs in order to maximize value.
measuring value of information systems, operational efficiency, productivity gains due to information systems.

43. Bannister, G.; Stolp, C. European Journal of Operational Research 1995, 80(3), 672-691.

44. Barr, R. S.; Seiford, L. M.; Siems, T. F. An Envelopment-Analysis Approach to Measuring the Managerial Efficiency of Banks 1992.

45. Bauer, P. W. Journal of Econometrics 1990, 46, 39-56.
A number of techniques have been developed that expand the range of options available to researchers for estimating the frontiers. This paper discusses recent developments in the econometric approach to the estimation of stochastic frontiers such as production, costs, and profit functions. Areas requiring further work are also noted.

46. Beasley, J. E. Omega 1990, 18( 2), 171-83.
The author presents a quantitative model for comparing university departments concerned with the same discipline. This model is based upon ideas drawn from data envelopment analysis. Computational results are given for chemistry and physics departments in the United Kingdom.
Education. Operations research.

47. Belton, V.; Vickers, S. P. Journal of the Operational Research Society 1993, 44( 9), 883-96.
Both data envelopment analysis (DEA) and multiple criteria analysis (MCA) can be used to assess the efficiency with which units perform similar tasks. This paper describes an approach derived from the integration of data envelopment analysis and a multi-attribute value function. This approach is implemented as a visual interactive decision support system, the use of which is illustrated by a practical application. The authors feel that this approach overcomes some of the limitations of the original DEA approach and, in particular, increases users' understanding of DEA. This approach is particularly well suited to the analysis of the efficiency of a small number of units.
decision support systems. operations research.

48. Berger A.; Haweck G.; Humphrey D. Journal of Monetary Economics 1987, 20, 501-520.
Conventional scale and scope economies are inadequate to determine the competitive viability of banks that vary in scale and product mix simultaneously. The authors develop two new methods that deal with this phenomenon ad they also apply the methods to banking data.

49. Berger A.; Leusner J.; Mingo J. Georgia Workshop on Productivity Measurement 1994.
The study assesses the technical efficiency of a large network of bank branches over three year period. The authors employe a trigonometric cost function which proves to be more adequate from the translog one.

50. Bessent, A.; Bessent, W. Educational Administration Quarterly 1980, 16(2), 57-75.

51. Bessent, A.; Bessent, W. A Fractional Programming Model for Determining the Efficiency of Decision Making Units EFIC Clearinghouse on Education Management, University of Oregon, Eugene Oregon: 1981.

52. Bessent, A.; Bessent, W. Productivity in Community Colleges: a Technique for Determining Relative Efficiency Community College Productivity Center, Dallas, TX: 1981.

53. Bessent, A.; Bessent, W.; Charnes, A.; Cooper, W. W.; Thorogood, N. Educational Administration Quarterly 1983, 19 (2), 82-107.

54. Bessent, A.; Bessent, W.; Elam, J.; Clark, T. Interfaces 1984, 14(6), 1-8.

55. Bessent, A.; Bessent, W.; Elam, J.; Clark, T. Operations Research 1988, 36( 5), 785-96.
Assessing the efficiency of organizational units has long been recognized as an important management problem but an extremely difficult one to solve, especially when one considers the multiplicity of resources and services or products associated with these units operating in a variety of industries. Data envelopment analysis (DEA), a promising approach for addressing this problem, is extended in the paper. The proposed extension, called constrained facet analysis (CFA), provides a lower bound efficiency measure for organizational units that have a mix of resources and/or outputs that is different from any frontier point. CFA yields marginal rates of productivity and substitution associated with the lower bound efficiency measure.
Management science. Operations research.

56. Bessent, A.; Bessent, W.; Kennington, J.; Regan, B. Management Science 1982, 28(12), 1355-1367.

57. Bitran, G. R.; Valor-Sabatier, J. Advances in Mathematical Programming and Financial Planning 1987, 1, 61-84.

58. Bjurek, H.; Hjalmmarsson, L.; Forsund, F. R. Journal of Econometrics 1990, 46, 213-227.
The purpose of this study is to analyse productive efficiency in about 400 local social insurance offices of the Swedish social insurance system for the period 1974-1984. The analysis is based on parametric and nonparametric deterministic frontiers. The general results are that the efficiency is around 0.8 and that the differences between the approaches are surprisingly small.

59. Borden, J. P. Journal of Accounting and Public Policy 1988, 7(2), 77-96.

60. Boussofiane, A.; Dyson, R. G.; Thanassoulis, E. European Journal of Operational Research 1991, 52( 1), 1-15.
Data envelopment analysis (DEA) a linear programming based technique for measuring the relative performance of organisational units where the presence of multiple inputs and outputs makes comparison difficult. The paper introduces the technique and focuses on some of the key issues that arise in applying DEA in practice.
Linear programming. Operations research.

61. Bowen, W. M. Computers, Environment and Urban Systems 1990, 14( 2), 133-44.
Using subjective judgements as reflected in an analytic hierarchy process (AHP) of site selection, a comparison is made to an objective data envelopment analysis (DEA) selection procedure. Similarity in the results are discussed and conditions for expectations of parallel and different results are explored. A two step process of integrating subjective and objective decision making in site selection is posited.
Decision support systems. Decision theory. Linear programming. Operations research. Town and country planning.

62. Bowlin, W. F. The Government Accountants Journal 1986, 35(2), 50-57.

63. Bowlin, W. F. J. Oper. Res. Soc. (GB) 1987, 38( 2), 127-35.
Evaluates a new methodology, data envelopment analysis (DEA), as a tool for measuring and evaluating the operational efficiency of US Air Force organizations. The study involves the application of DEA to locate possible inefficiencies in the performance of US Air Force real-property maintenance activities. Dynamic behaviour, as exhibited via window analyses is reviewed along with static counterparts.
data analysis. linear programming. military computing.

64. Bowlin, W. F. JAI Press: 1989; Vol. 5.
Data Envelopment Analysis (DEA) provides information on sources and amounts of inefficiences and distinguishes between technical and scale efficiencies which suggest that DEA has the potential to be a viable tool for economy and efficiency audits. This study investigates the applicability and feasibility of using DEA for such audits of U.S. Air Force base level accounting and finance offices (AFOs). Specifically, technical and scale efficiencies were measured from 1983 through 1985. Actions taken to improve the technical efficiency of AFOs were effective in 1984, but not in 1985.

65. Bowlin, W. F.; Charnes, A.; Cooper, W. W.; Sherman, H. D. 1985; Vol. 2.

66. Bowlin, W. F.; Wallace, J. R. I.; Murphy, R. L. Efficiency-Based Budgeting Forthcoming.

67. Boyd; Fare, R. European Journal of Operational Research 1984, 15, 331-332.

68. Burgess, J. Management Science (Forthcomming) .
Recent emphasis inthe economci theory of organisations has focused on the relative degree of contract failure in controlling and monitoring economic activity to determine comparative advantages in efficiency between different ownrship structures. This study analyses four types of ownrship strucutre in the US hospital industry-private nonprofit, private for-profit, federal and state and local government. Distance function are used to measure technical efficiency of hospitals producing multiple outputs relative to other hospitals in the sample. Hospitals with different ownership types are compared in terms of technical efficiency, scale efficiency, and other factors.

69. Burgess, J.; Wilson, P. Journal of Productivity Analysis .
Federal and non-federal hospitals are assessed for their technical efficiency. Tecnological change is compared and also significant cost efficiencies are found by comparing the two types of hospitals.

70. Burgess, J.; Wilson, P. Working Paper 1994.
This study analyses annual data from federal and nonfederal hospitals in the period 1985-1988. The use of panel data regression techniques on measured technical inefficiency allows for determinantns of technical inefficiency out of the control of hospital managers to be identified and measured over time.

71. Byrnes, P.; Fare, R.; Grosskopf, S. Management Science 1984, 30(6), 671-681.

72. Byrnes, P.; Fare, R.; Grosskopf, S.; Lovell, C. A. K. Management Science 1988, 34(9), 1037-1053.

73. Byrnes, P.; Grosskopf, S.; Hayes, K. Review of Economics and Statistics 1986, 65, 337-341.

74. Callen, J. L.; Falk, H. The Accounting Review 1993 , 68(1), 48-65.
This article relates the efficiency of nonprofit organizations to the composition of their board of trustees. Following arguments by Fama and Jensen (1983a, 1983b) and Williamson (1983), we conjecture that nonprofit organizations are more efficient if their board of trustees have a larger proportion of outsider trustees rather than insider (employee) trustees since the former are presumed to have a greater incentive to monitor the organization and the latter are presumed to have a greater incentive to consume perquisitesss. This conjecture is tested empirically for a sample of 72 charities with a specific health focus, with respect to both the technical and allocative efficiency of the organization.

Technical efficiency indices were derived by using data envelopment analysis (DEA) under various assumptions concerning the industry production technology, including the assumption that charity outputs are not substitutable for each other. These indices are reasonably well measured since the data included information on volunteer labor, a necessary input for charity services.

To test the conjecture that the technical efficiency of th organization is affected by the composition of the board of trustees, the estimated technical indices of technical efficiency were regressed on the board's proportion of insider trustees and the organization's debt-value ratio. The conjectured relationship was not confirmed by the data. As an additional test, the proportion of insider trustees was replaced by the ratio of insider-trusstess remuneration to total labor remuneration, the argument being that the greater the relative remuneration received by the insider trustees, the more capable they bight be in appropriating perquisites of all types. As before, the results did not confirm the conjectured relationship, leading us to conclude that the the technical efficiency of nonprofits is not affected by the composition of the board of trustees.

To determine the relationship, if any, between the allocative efficiency of the organization and the composition of the board of trustees, charity costs were multiplied by the technical efficiency index so that any remaining cost inefficiencies are allocative in nature. This adjusted cost was then regressed on the proportion of insider trustees on the board (or the proportion of insider-trustee remuneration) and the organization's debt-value ratio. The conjectured relationship between allocative efficiency and the composition of the board of trustees was also not confirmed by the data.
nonprofit organizations, efficiency, board of trustees, data envelopment analysis.

75. Camm, J. D.; Grogan, T. J. Interfaces 1988, 18( 6), 52-60.
Economists have used frontier analysis for some time now to measure the efficiency of the firm. Recently, particularly since the development of data envelopment analysis, management scientists have applied efficient frontier analysis to a variety of problems. The authors used linear programming to estimate a frontier function that describes where the age records for a particular running race should be, given the current records. The frontier function is used to handicap running races based on age and sex. The frontier method is an improvement over current methods in that it avoids the problem of soft records for some ages.
Entertainment. Linear programming.

76. Capettini, R.; Dittman, D. A.; Morey, R. C. Journal of Accounting and Public Policy 1985, 4(2), 83-110.

77. Cave, M.; Hanney, S.; Kogan, M.; G., T. Jessica Kingsley Publishers Ltd.: London, 1988.

78. Chan, P. S.; Sueyoshi, T. International Journal of Systems Science 1991, 22( 9), 1625-36.
The impact of a significant environmental change, namely, deregulation, on strategic management and performance in the US airline industry is examined. Firm performance is deemed to be affected by the configuration of the competitive environment, strategy and structure. Using the methodology of data development analysis (DEA), airline performance before and after deregulation is compared. The results support the gestalt proposition that the configurations under regulation differ significantly from those under deregulation. The promising use of DEA in strategic management research is underlined.
Commerce. Travel industry.

79. Chang, K.-P.; Guh, Y.-Y. European Journal of Operational Research 1991, 52( 2), 215-23.
Shows that the data envelopment analysis (or Farrell) method is not nonparametric in the sense that the method implicitly employs linear production functions (frontiers) to measure efficiency. Deriving positive-multiplier linear production frontiers and grouping the decision making units (DMUs) by their corresponding frontiers can simplify the estimation procedures when there are new DMUs included, and can easily determine how many inputs should be reduced or/and how many outputs should be increased to make an inefficient DMU efficient. This procedure can also circumvent the problems of the non-Archimedean infinitesimal in the Charnes-Cooper-Rhode (1978, 1979) literature. It is also shown that, because of the linear production functions, both Banker's (1984) and the Banker-Charnes-Cooper methods (1984) fail in testing for returns-to-scale.
Management science. Production control.

80. Chang, K.-P.; Kao, P.-H. Kluwer Academic Publishers: Dordrecht, 1992.
The articl has employed the data envelopment analysis method to evaluate the efficiency of the five bus firms in Taipei city. When vehicle kilometers (revernue or the measure combining vehicle kilometers, revenue and the number of traffic trips on routes) was used as the output measure, it concluded that the publicly owned Taipei Municipal Bus had increased (not increased) it technical efficiency after the government liberalized the urban bus market. This article also found that in both the one output (vehicle kilometers) and three outputs cases, Taipei Municipal Bus had, on an average, lower efficiency scores than the private firms, and that while each firm usually employed a linear production technology for several consecutive years the private firms were more flexible in adopting different technologies.
DEA.

81. Chang, Y.-L.; Sueyoshi, T. Computer Science in Economics and Management 1991, 4( 1), 51-64.
The article describes a general-purpose microcomputer code for data envelopment analysis (DEA) that incorporates four different DEA models in the form of a user-friendly, menu-driven structure.
Data analysis. Mathematics computing. Microcomputer applications. Statistical analysis.

82. Charnes, A.; Clark, T.; Cooper, W. W.; Golany, B. 1985; Vol. 2.

83. Charnes, A.; Clarke, R.; Cooper, W. W. JAI Press: 1989; Vol. 5.

84. Charnes, A.; Clarke, R.; Divine, D.; Ruefli, T. W.; Thomas, D. JAI Press: 1989; Vol. 5.

85. Charnes, A.; Cooper, W. W. Accounting, Organizations and Society 1980, 5(1), 87-107.

86. Charnes, A.; Cooper, W. W. Journal of Enterprise Management 1980, 2(2), 143-162.

87. Charnes, A.; Cooper, W. W. Journal of Enterprise Management 1978, 1(1), 5-23.

88. Charnes, A.; Cooper, W. W. European Journal of Operational Research 1984, 15(3), 333-334.

89. Charnes, A.; Cooper, W. W. 1985; Vol. 2.

90. Charnes, A.; Cooper, W. W.; Golany, B.; Seiford, L.; Stutz, J. Journal of Econometrics 1985, 30(1/2), 91-107.

91. Charnes, A.; Cooper, W. W.; Huang, Z. M.; D.B., S. Journal of Econometrics 1990.

92. Charnes, A.; Cooper, W. W.; Huang, Z. M.; Sun, D. B. International Journal of Systems Science 1991, 22( 11), 2057-77.
Analytic relations between the two representations (half-space and finitely generated) of cones in polyhedral cone-ratio DEA (data envelopment analysis) models are derived, which, for useful classes of cases in which they are simple, can make available the advantages of both for analysis and computation. The authors prove that such transfers between the two may be made through, essentially, the inverse of an associated matrix available in the computation and provide formulae connecting solutions in the two representations.
Combinatorial mathematics. Matrix algebra.

93. Charnes, A.; Cooper, W. W.; Learner, D. B.; Phillips, F. Y. Journal of Marketing 1985, 49(3).

94. Charnes, A.; Cooper, W. W.; Lewin, A. Y.; Morey, R. C.; Rousseau, J. 1985; Vol. 2.

95. Charnes, A.; Cooper, W. W.; Lewin, A. Y.; Seiford, L. M. Kluwer: 1995.

96. Charnes, A.; Cooper, W. W.; Li, S. Soci-Economic Planning Sciences 1989.

97. Charnes, A.; Cooper, W. W.; Niehaus, R.; Shinnar, A. Work, Organizations, and Technological Change; NATO Conference Series, Series II: Systems Science; Plenum Press: pp 2414-2456.

98. Charnes, A.; Cooper, W. W.; Qei, Q. L.; Huang, Z. M. International Journal of Systems Science 1989, 20( 7), 1099-118.
A new 'cone ratio' data envelopment analysis model that substantially generalizes the Charnes-Cooper-Rhodes model and the Charnes-Cooper-Thrall approach characterizing its efficiency classes is developed and studied. It allows for infinitely many decision-making units and arbitrary closed convex cones for the virtual multipliers as well as the cone of positivity of the vectors involved. Generalizations of linear programming and polar cone dualisations are the analytical vehicles employed.
Decision theory. Linear programming.

99. Charnes, A.; Cooper, W. W.; Rhodes, E. Journal of Enterprise Management 1980, 2(2), 160-162.

100. Charnes, A.; Cooper, W. W.; Rhodes, E. Management Science 1981, 27(6), 668-697.

101. Charnes, A.; Cooper, W. W.; Rhodes, E. European Journal of Operations Research 1978, 2(6), 429-44.
A nonlinear (nonconvex) programming model provides a new definition of efficiency for use in evaluationg activities of no-for-profit entities participating in public programs. A scalar measu8re of the efficiency of each participating unit is thereby provided along with methods for objectively determining weights by reference to the observational data for the multiple outputs and multiple inputs that characterize such programs. Equivalencies are estalished to ordinary linear programming models for effecting computations. The duals to these linear programming models provide a new way for estimating extremal relations from observational data. Connections between engineering and economic approaches to efficiency are delineated along with new interpretations and ways of using them in evaluating and controlling managerial behavior in public programs.

102. Charnes, A.; Cooper, W. W.; Rhodes, E. European Journal of Operational Research 1979, 3(4), 339.

103. Charnes, A.; Cooper, W. W.; Schinnar, A. Omega 1982, 10(2).

104. Charnes, A.; Cooper, W. W.; Seiford, L.; Stutz, J. Operations Research Letters 1983, 2(3), 101-103.

105. Charnes, A.; Cooper, W. W.; Seiford, L.; Stutz, J. Socio-Economic Planning Sciences 1982, 16(5), 223-224.

106. Charnes, A.; Cooper, W. W.; Thrall, R. M. Oper. Res. Lett. (Netherlands) 1986, 5( 3), 105-10.
DEA (data envelopment analysis) attempts to identify sources and estimate amounts of inefficiencies contained in the outputs and inputs generated by managed entities called DMUs (decision making units). Explicit formulation of underlying functional relations with specified parametric forms relating inputs to outputs is not required. An overall (scalar) measure of efficiency is also obtained for each DMU from the observed values of its multiple inputs and output without requiring uses of a priori weights. There are many different ways of specifying DEA reference sets. A partition into six classes is provided for such observations in which three are scale inefficient and three are scale efficient with the latter containing subsets of DMUs that are also technically (=zero waste) efficient.
linear programming. statistical analysis.

107. Charnes, A.; Haag, S.; Jaska, P.; Semple, J. International Journal of Systems Science 1992, 23( 5), 789-98.
In contrast to existing sufficient conditions for preservation of efficiency under special perturbations and matrix structural assumptions, sensitivity of the additive model's classifications in data envelopment analysis (DEA) is investigated by means of new DEA formulations focusing on the stability (sensitivity) of an organization's classification (whether efficient or inefficient). The formulations for the additive model are linear programming problems whose solutions yield a particular region of stability, a 'cell', in which an organization's classification remains unchanged.
Linear programming. Management science. Sensitivity analysis.

108. Charnes, A.; Neralic, L. European Journal of Operational Research 1990, 48( 3), 332-41.
Sensitivity analysis of the additive model in data envelopment analysis is studied in the paper. Sufficient conditions for simultaneous change of all outputs and all inputs of an efficient decision making unit which preserves efficiency are established. A numerical illustration is provided.
Decision theory. Linear programming.

109. Charnes, A.; Rousseau, J. J.; Semple, J. H. International Journal of Systems Science 1992, 23( 12), 2401-6.
Two problems in linear programming associated with data envelopment analysis (DEA) namely, employing nonarchimedean infinitesimals, transcendentals ('big Ms') and categorical variables (in a new nonarchimedian formulation) are addressed. A new more sophisticated pricing procedure as part of an adjacent extreme point algorithm solves these efficiently in the base field. Employing this in Charne's nonarchimedean simplex algorithm led to a ninefold increase in computational speed on large DEA problems with about 1000 decision-making units. Additionally, computational failures due to cycling and/or conditioning instabilities were eliminated.
costing. economic cybernetics. linear programming.

110. Charnes, A.; Rousseau, R. R.; Semple, J. 677; An Effective Non-Archimedean Anti-Degeneracy/Cycling Linear Programming Method Especially for Data Envelopment Analysis and Like Methods Center for Cybernetic Studies: 1992.
A non-Archimedean effective anti-degeneracy/cycling method for linear programming m odels, especially data envelopment analysis (DEA), processing networks, and advertising media mix models is herein developed. It has given a tenfold speed increase plus elimination of cycling difficulties over conventional Marsden or Kennington/Ali software modules ina 1,000 LP DEA application.
degeneracy/ cycling, non-Archimedean data envelopment analysis, linear programming.

111. Charnes, A.; Rousseau, R. R.; Semple, J. Sensitivity and Stability of Efficiency Classifications in Data Envelopment Analysis 1993Forthcoming.
A new technique for assessing the sensitivity and stability of efficiency classifications in data envelopment analysis (DEA) is presented. Here developed for the ratio (CCR) model, this technique extends easily to other DEA variants. An organization's input-output vector serves as the center for a (symmetric) cell within which the organization's classification remains unchanged under perturbations of the data. The size and shape of the cell are determined by the vector norm chosen. For the L1, L-infinity, and generalized L-infinity norms, the largest such cell can be computed using linear programming formulations. The size of the cell can be interpreted as a measure of the classification's stability, especially with regards to errors in the data.

112. Charnes, A.; Semple, J. Sensitivity of Efficiency Classifications in the Ratio Model of Data Envelopment Analysis 1992Forthcoming.
A new technique for assessing the sensitivity and stability of efficiency classifications in data envelopment analysis (DEA) is presented. An organization's input-output vector serves as the center for a (symmetric) cell within which the organization's classification remains unchanged under perturbations of the data. The size and shape of the cell are determined by the vector norm chosen. For the L1, L-infinity, and generalized L-infinity norms, the largest such cell can be computed using linear programming formulations. The size of the cell can be interpreted as a measure of the classification's stability.

113. Charnes, A.; Zlobec, S. ZOR, Methods and Models of Operations Research 1989, 33( 3), 167-79.
Efficiency evaluations in data envelopment analysis are shown to be stable for arbitrary perturbations in the convex hulls of input and output data. Also, the corresponding restricted Lagrange multiplier functions are shown to be continuous. The results are proved using point-to-set mappings and a particular region of stability from input optimization.
Operations research. Set theory. Stability.

114. Chilingerian J.; Sherman D. Annals of Operations Research 1995.
The authors provide a combined DEA approach for assessing physicians efficiency. Weight restrictions are proposed as follows: Assess the DMUs and select from those that were found efficient a subset that satisfies certain criteria. Derive weight restrictions from this subset and reassess the original data set. Problems of target setting are reported in the results.

115. Chu, X.; Fielding, G. J. Transportation Research, Part A (Policy and Practice) 1992, 26A( 3), 223-30.
Uses data envelopment analysis (DEA) to develop a single measure for the efficiency and a single measure for the effectiveness of a transit agency relative to other agencies within the same peer group. By using a single measure for each of these criteria, the paper provides a more robust indicator of transit performance than the widely used multiple ratio analysis performed in the Irvine performance evaluation method (IPEM). The DEA model is applied to two transit agency peer groups-one serving large metropolitan areas and the other serving relatively small cities and large towns.
Mathematical programming. Operations research. Transportation.

116. Clarke, R. L. Decision Sciences 1992, 23( 2), 376-84.
In response to increasing costs and reductions in manpower, the Tactical Air Command (TAC) of the USAF experimented with a specialized productivity measurement model known as data envelopment analysis (DEA). A medium-sized application of DEA was employed by TAC to evaluate the productivity of its seventeen subordinate vehicle maintenance sections over a four-year period. The application reports gains in productivity and the reactions of the field managers to the use of DEA.
linear programming. maintenance engineering. transportation.

117. Cook, W. D.; Johnston, D. A.; McCutcheon, D. OMEGA International Journal of Management Science 1992, 20(2), 227-239.
This paper examines the implementation of robotics in 31 companies. The problem is to determine which companies are the most efficient implementators. Implementation efficiency is modelled as a multiple criteria problem, using DEA. This tool was designed for evaluating the relative efficiency of a set of decision making units in the presentce of multiple inputs and multiple outputs. In the present setting, three outputs: starting time, uptime when required, and management satisfaction are examined relative to three inputs or circumstances: system complexity, system novelty, and the implementation team's previous experience with the technologies involved. In addition, three variables, namely supplier management practices, plant size, and perceived urgency are evaluated vis-a-vis their impact on implementation efficiency.
implementation, robotics, multicriteria, efficiency, DEA.

118. Cook, W. D.; Kress, M. Management Science 1990 , 36( 11), 1302-10.
Presents a general model for aggregating votes from a preferential ballot. The thrust of the model is to accord each candidate a fair assessment in terms of his overall standing vis-a-vis first place, second place, . . . kth place votes. The form of the model is a combined index Sigma /sup k//sub j=1/W/sub j/ nu /sub ij/ is the number of the jth place votes received by the ith candidate. The weights W/sub j/ are assumed to form a monotonically decreasing sequence with W/sub j/-W/sub j+1/>or=d(j, in ). These constraints correspond to the assurance region (AR) side constraints in the data envelopment analysis framework. The properties of the model are examined in terms of the discrimination intensity function d, and in the special case that d(j, in )= in , the model is shown to be equivalent to the consensus models of Borda (1781) and Kendall (1962).
Behavioural sciences. Management science.

119. Cook, W. D.; Kress, M. European Journal of Operational Research 1991, 54( 2), 191-8.
Develops a model for aggregating ordinal preference rankings across multiple criteria. Criteria weights are developed as part of the optimization process. The model utilizes three levels of ordinal priorities: (1) preferences among alternatives on the basis of each criterion; (2) preferences among the criteria to reflect the importance of those criteria; and (3) preferences among the criteria which express the degree to which the respondent can clearly distinguish among the alternatives. The model is based upon a data envelopment analysis structure which permits a flexible weighting mechanism for each alternative being ranked.
Decision theory. Management science. Optimisation.

120. Cook, W. D.; Kress, M.; Seiford, L. M. Journal of the Operational Research Society 1993, 44( 2), 133-40.
In many problems involving efficiency analysis using data envelopment analysis (DEA), certain factors may be measurable only on an ordinal scale. Specifically, it may be possible only to rank order the decision making units (DMUs) according to a factor, rather than being able to assign a specific numerical value of that factor to each DMU. To illustrate this, the authors examine a problem involving the evaluation of new technology installations. The presence of qualitative factors in such an environment motivates the need to investigate how such factors can be incorporated into existing efficiency measurement models. In particular, a procedure is presented for incorporating an ordinal factor into the DEA structure, with the resulting formulation being a particular form of cone ratio model. The model is then applied to the technology installation efficiency problem.
decision theory. linear programming.

121. Cook, W. D.; Kress, M.; Seiford, L. M. European Journal of Operational Research 1992, 59( 2), 319-23.
Data envelopment analysis (DEA) has received significant attention as a tool for measuring the relative efficiency of each member of a set of decision making units (DMUs). Typically, a relatively large proportion of the DMUs will be credited with an efficiency score of 1, with no clear means of discriminating among such units. In a number of applications, however, it may be necessary to select a 'winning' DMU from this set of frontier units. The paper examines various conditions that are imposed on the multipliers in a DEA analysis. In each case, an approach is suggested for breaking ties on the frontier.
decision theory. linear programming.

122. Cook, W. D.; Roll, Y.; Kazakov, A. INFOR 1990 , 28( 2), 113-24.
A pilot application of data envelopment analysis (DEA) for the measurement of the efficiency of highway maintenance patrols is demonstrated. Selection of pertinent factors is discussed and the potential benefits of the analysis listed. A bounded DEA model is constructed and results are compared to those obtained from a conventional DEA model. The effects of secondary factors on the relative efficiencies of patrols are examined by analyses of sub-groups of decision making units (DMUs), differing in the intensities of the respective factors.
Decision theory. Linear programming. Operations research. Transportation.

123. Cooper, W. W.; Huang, Z.; Li, S. X. Stochastic Models in Data Envelopment Analysis 1993Forthcoming.
New stochastic data envelopment analysis methods are introduced and developed in this paper. Necessary and sufficient conditions for a DMU to be stochastically efficiency are derived. By introducing the evaluator's risk and relaxing the group stochastic efficiency definitions, we provide an approach to evaluate DMU's stochastic efficiencies by solving a series of almost 100% assurance chance constrined programming problems. Under som special assumptions on the joint probability distribution of th input and output data, we can reduce chance constrained programming problems to regular (convex) nonlinear programming problems.
risk, stochastic efficiency, stochastic frontier, almost 100$ assurance chance constrained programming, nonlinear programming.

124. Deprins, D.; Simar, L.; Tulkens, H. Elsevier Science Publishers B.V.: North-Holland, 1984.

125. Desai, A.; Storbeck, J. E. Computers, Environment and Urban Systems 1990, 14( 2), 145-56.
A new technique for measuring relative spatial efficiency is proposed. A data envelopment analysis (DEA) model for evaluating the efficiency of incremental siting decisions is formulated. The model provides an intuitively attractive approach to measuring accessibility on the basis of multiple spatial criteria, and articulates a ratio-scale based understanding of the 'noninferiority' of locational decisions. As such, it puts the more useful aspects of the multiobjective siting process within the grasp of the decision maker. This work rests squarely on the conceptual foundation laid down by H. Fisher and G. Rushton (1979). Consequently, the model might best be viewed as a first attempt to operationalize their constructs.
Decision support systems. Decision theory. Operations research. Optimisation. Town and country planning.

126. Desai, A.; Walters, L. C. Decision Sciences 1991 , 22( 2), 335-53.
The conversion of multivariate data analysis into information that managers can readily use in their decision-making has always been a challenging task. The authors provide a parallel axis representation of multivariate data and demonstrate how this representation may be used as a tool for implementing the results of data envelopment analysis. They use an analysis of the efficiency of electricity generating plants to illustrate the technique.
Management science. Production control.

127. Dey, D.; Seidmann, A. Information Systems Research 1994, 5(3), 275-293.

128. Dogramaci, A.; Fare, R. Kluwer Academic Publishers: Boston, 1988.

129. Doukas J.; Switzer L. Journal of International Financial MarketsInstitutions and Money 1991, 1(2).
This is an econometric paper that investigates economies of scale and scope of 563 bank branches in Canada. The authors found evidence of economies of scale but not strong indications of economies of scope.

130. Doyle, J. R.; Green, R. H. Omega 1991, 19( 6), 631-8.
Using an application of linear programming known as data envelopment analysis (DEA), a method is described to compare products which vary in excellence along a number of dimensions, and for each of which there might be a number of associated 'costs'. The method is illustrated by comparing published benchmarks of 37 computer printers. Potential uses of a DEA analysis of products might be: to assist corporate buyers who may need to reconcile a diversity of present and future uses in one standardised purchase; in competitor analysis; in determining unexplored market niches; and as a normative model of product excellence against which product purchasing behaviour could be compared.
Linear programming. Management science.

131. Doyle, J.; Green, R. OMEGA International Journal of Management Science 1993, 21(6), 713-715.
Stewart has recently presented a survey of theory and practice in multiple criteria decision making (MDCM). WHile Sterwart admits that his is "by no means a comprehensive review.

132. Doyle, J.; Green, R. Journal of the Operational Research Society 1994, 45(5), 567-578.
Examines a neglected aspect of data envelopment analysis: cross-efficiency. The authors develop the concept of cross-effiency in a number of new directions. The cross-efficiency in the concept of peer-appraisal, as opposed to self-appraisal implied by simple efficiency, and discuss the relative merits of each. The authors also present mathematical formulations of, and intuitive meanings for three possible implementations of aggressive and benevolent cross-efficiency. The authors have implemented two of these formulations in computer programs; their performance is compared empirically on a real data set. Finally, the authors suggest practical uses for cross-efficiency, illustrated with reference to the same data set.
decision theory, linear programming.

133. Doyle, J.; Green, R. Journal of Information Technology 1994, 9(1), 61-69.
A linear programming approach (data envelopment analysis) is described to determine the relative merits of a set of multi-input, multi-output systems, in which more output for less input is considered good. The method is applied to benchmarks of microcomputers, and is contrasted with a multiple regression analysis of the same data. It is also argued that the essence of two opposing strategic outlooks can be captured within the method.
linear programming, microcomputer, performance evaluation.

134. Drake, L.; Howcroft, B. OMEGA International Journal of Management Science 1994, 22(1), 83-90.
This paper investigates the relative efficiency of a UK clearing bank's branches using the nonparametric programming methodology. In addition to calculating indices of technical (productive) efficiency for all branches in the sample, this measure of technical efficiency is dichotomized into its constituent components, scale efficiency and technical efficiency. Finally, the non-parametric technique is utilized to investigate the causes of observed inefficiency in the case of one illustrative branch example in detail.
DEA, technical efficiency, scale efficiency, banking, branch networks.

135. Duncombe W. The Review of Economics and Statistics 1992, 180-184.
Evidence on costs and factor substitution are presented for a sample of local fire departments in New York. The Leontief, Cobb-Douglas and CES technologies have been rejected. Significant effects from socioeconomic factors have also been identified.

136. Durchholz, M.; Barr, R. Annals of Operations Research 1995, (forthcoming).

137. Dyson, R. G.; Thanassoulis, E. Journal of the Operational Research Society 1988, 39( 6), 563-76.
Explores the consequences of total weight flexibility in data envelopment analysis (DEA) assessments of the efficiency of decision-making units, and it suggests one possible way of limiting such flexibility. It is suggested that total weight flexibility can lead to some decision-making units being assessed, in effect, on only a small subset of their inputs and outputs, while the rest of their inputs and outputs are all but ignored. Constraining the weights in DEA assessments overcomes this problem. The paper suggests one way in which constraints can be placed for the case where the decision-making units to be assessed use only a single input. The method is illustrated using data on local-authority rates departments. Finally, the paper discusses the interpretation and usefulness of the information obtained from DEA assessments involving weights constraints.
Decision theory. Public administration.

138. Elam, J. J. Public Productivity Review 1989, XII(3), 263-277.

139. Epstein, M. K.; Henderson, J. C. Decision Sciences 1989, 20( 1), 90-119.
Data envelopment analysis (DEA) is a linear programming-based technique that converts multiple input and output measures into a single comprehensive measure of productive efficiency. The authors draw on theories of decision making, measurement and control, the mathematical properties of DEA, prior reported applications, and their own experience, to assess the potential of DEA as a general management tool. They first make the distinction between managerial diagnosis and control, and show how measurement requirements differ for these two managerial decision contexts, and then argue that DEA has the potential to provide support in each context.
Decision support systems. Linear programming. Management information systems. Performance evaluation.

140. Fare, R. Operations Research Letters 1986, 5(2), 83-85.

141. Fare R.; Grosskopf, S. European Journal of Operational Research 1994, 79, 379-382.
Fare makes a points that his approach for assessing economies of scale is similar to the one suggested by Banker and the rest.

142. Fare, R.; Grosskopf, S. Zeitschrift Fur Nationalokonomie 1983, 43(3), 257-271.

143. Fare, R.; Grosskopf, S. European Journal of Operational Research 1983, 13, 173-179.

144. Fare, R.; Grosskopf, S.; Logan, J. Resources and Energy 1985, 5(4), 349-367.

145. Fare, R.; Grosskopf, S.; Lovell, C. A. K. Kuwer Academic Publishers: Boston, 1985.

146. Fare, R.; Grosskopf, S.; Njinkeu, D. Management Science 1988, 34(12), 1507-1511.

147. Fare, R.; Hunsaker, W. Management Science 1986 , 32(2), 237-243.

148. Ferrier, G.; Kerstens, K.; Vanden Eeckaut, P. CORE Discussion Paper 9423 .
They compare alternative radial and non-radial models for their efficiency values.

149. Forsund, F. R.; Hjalmarsson, L. Analyses of Industrial Structure: a Putty-Clay Approach The Industrial Institute for Economic and Social Research, Stockholm, Sweden: 1987.

150. Forsund, F. R. Kluwer Academic Publishers: Dordrecht, 1992.
The trunk road system in Norway has to be supplemented by a number of ferries due to the long coastline with numerous islands and fjords. Most of the ferries are run by private companies, but at a loss. The deficits are covered by the Ministry of Transport. The subsidies have risen rapidly in the last years and have focussed attention on whether the ferries are really run as efficiently as possible. To change the incentives to economize, a lump-sum payment is considered. To implement such a system, an initial assessment of reasonable input requirements is needed. The aim of this article is to provide such a yardstick by establishing a best practice frontier. Both a non-parametric and parametric approach to a deterministic frontier are tried and differences of results discussed. Peculiarities due to a choice of methods are revealed. The efficiency distributions are quite similar for the two methods except for scale efficiency, where the paratmetric method indicates substantial unrealized scale economies, while the non-parametric approach shows the largest and some small ferries to be scale efficient. The results indicate a substantial rationalization potential of about 30 percent in total.
DEA.

151. Fox, W.; Hofler, R. Southern Economic Journal 1986 , 53(2), 461-477.
A two output multiple input production technoclogy is employed by the authors using homothetic functions. Allowance is given to both techical and allocative inefficiencies. The methodology is aplied on panel data from 176 water utilities. The model uses two outputs, two inputs, production condition variables, quality variables and prices. Chow tests were also employed to test efficiency differences due to different ownership.

152. Frank, R. G. Management Science 1988, 34 (8), 1026-1029.

153. Giokas, D. I. Omega 1991, 19( 6), 549-57.
A comparison regarding the operational efficiency of individual branches of a bank is made, through the application to the same body of data of two different estimation methods: data envelopment analysis; and loglinear model analysis. In addition to that, the study examines whether operations in the bank branches were conducted in regions of increasing, constant or decreasing returns to scale. The DEA results suggest that increasing, constant or decreasing returns to scale may be observed in different regions of the production function, whereas the loglinear model suggests that increasing returns to scale are in operation.
Banking. Mathematical programming.

154. Golany, B. Journal of the Operational Research Society 1988, 39( 8), 725-34.
Presents a new, interactive multi-objective linear-programming procedure to aid decision-makers in setting up goals for desired outputs. The procedure relies on empirical production functions generated by the use of data envelopment analysis. It presents the decision-maker with a set of alternative efficient points in order either to compare different sets of inputs in terms of their effectiveness for goal achievement, or to set goals against which future management performance may be measured. With each iteration the new information provided by the decision-maker is used to adjust the procedure, leading to points which have greater effectiveness utility for the decision-maker. A numerical example is provided along with guidelines for future applications.
Decision theory. Linear programming. Operations research.

155. Golany, B. Management Science 1988, 34 ( 8), 1029-33.
Data envelopment analysis (DEA) evaluates the efficiency of decision making units without imposing a priori weights on the input and output dimensions involved. The paper extends DEA to situations where ordinal relations among weights corresponding to certain dimensions exist. The extended model is reduced to the original DEA formulation where some of the input and output factors are accumulated. An illustrative example is provided and potential areas of applications are pointed out.
Management science. Mathematical programming.

156. Golany, B.; Learner, D. B.; Phillips, F. Y.; Rousseau, J. J. Computers, Environment and Urban Systems 1990 , 14( 2), 89-102.
A description is given of a new productivity assessment based on data envelopment analysis (DEA) methodology. DEA offers the first satisfactory multi-input, multi-output measure of productivity, and allows for productivity management at the intrafirm and interfirm levels with particular application in marketing. Efficiency and effectiveness are distinguished as productivity components and means for managing each one is discussed and illustrated by an example from a marketing perspective.
DP management. Linear programming. Operations research. Production control. Service industries.

157. Golany, B.; Learner, D. B.; Phillips, F.; Rousseau, J. J. APORS '88: Selected Papers From the First Conference of the Association of Asian-Pacific Operational Research Societies Within IFORS.

158. Golany, B.; Phillips, F. Y.; Rousseau, J. J. IIE Transactions 1993, 25(6), 2-10.
Following the characterization via Data Envelopment Analysis (DEA) of managerial units as efficient or inefficient, management will wish to increase profitability and/or control control costs while becoming (or remaining) technically efficient in the DEA sense. This paper presents three families of models for achieving this and describes the managerial situations in which they are useful. The first addresses the management of an existing decision making unit (DMU) and the second attempts to identify the desired location for a new DMU. The third addresses the aggregate of all DMUs, reallocating scarce resources among them for maximum overall organizational profitability and technical efficiency.

159. Golany, B.; Roll, Y. Omega 1989, 17( 3), 237-50.
Data envelopment analysis (DEA) has become an accepted approach for assessing efficiency in a wide range of cases. The present paper suggests a systematic application procedure of the DEA methodology in its various stages. Attention is focused on the selection of 'decision making units' (DMUs) to enter the analysis as well as the choice and screening of factors. The application of several DEA models (in different versions and formulations) is demonstrated, in the process of determining relative efficiencies within the compared DMUs.
Decision theory. Management science.

160. Gong, B.-H.; Sickles, R. C. Journal of Econometrics 1992, 51, 259-284.
In recent years a mumber of alternative methods have been proposed with which to measure technical efficiency. However, we know little of their comparative performance. In this study we examine the relative strength of two different methodologies - stochastic frontier models (SF) and data envelopment analysis (DEA) - in estimating firm-specific technical efficiency. To address the limitations of previous studies, we utilize Monte Carlo techniques which allow us to control the structure of the underlying technology and the stochastic environment.

Most stochastic frontier models have focused on estimating average technical efficiency across all firms. The failure to estimate firm-specific technical efficiency has been regarded as a major limitation of previous stochastic frontier-models. To overcome this limitation we estimate firm-specific technical efficiency using panel data. We also examine the performance of stochastic frontier models using panel data for three estimators - maximum likelihood random effects, generalized least squares random effects, and within fixed effects.

Our results indicate that for simple underlying technologies the relative performance of the stochastic frontier models vis-a-vis DEA relies on the choice of functional forms. If the employed form is close to the given underlying technology, stochastic frontier models outperform DEA using a number of metrics. As the misspecification of the functional form becomes more serious and as the degree of correlatedness of inefficiency with regressors increases, DEA's appeal becomes more compelling. Our results also indicate that the preferred estimator for the SF model is the within estimator, which addresses two problems common to stochastic frontier models - the possible correlatedness of input levels and technical efficiency and the dependence of stochastic frontier models on distributional assumptions concerning the form of technical efficiency.

161. Greenberg, R.; Nunamaker, T. Financial Accountability and Management 1987, 3(4), 331-342.

162. Grosskopf, S.; Valdmanis, V. Journal of Health Economics 1987, 6, 89-107.

163. Gulledge, T. R. Jr.; Knox-Lovell, C. A. Kluwer Academic Publishers: Dordrecht, 1992.

164. Hansson, P.; Henrekson, M. Public Choice 1994 , 81, 381-401.
Does government spending have a positive or negative effect on economic growth ? The results of earlier empirical studies give mixed results. In this study we suggest a new method for testing the effect of different kinds of government expenditure on productivity growth in the private sector. The focus on productivity in the private sector and the useof disagreggated data makes it possible to avoid the mitigate a number of methodological problems. The major conclusions, which are quite robust, are that government tranfers, consumption and total outlays have consistently negative effects, while educational expenditure as a positive effect, and government investment has no effect on private productivity growth. The impact is also found to work solely through total factor productivity and not via the marginal productivity of capital and labour.

165. Hay, D. L. G. Institute of Economics and Statistics, Oxford 1995.

166. Hayes, R. D.; Millar, J. A. The Accounting Review 1993, 68(1), 89-92.
Ther "extension"of Hayes and Millar (1990) by Mensah and Li (1993) brings up essentially two areas of controversy. First, Mensah and Li find fault with our "implied"assumption of cross-sectionally fixed cost shares. Second, their use of an additional measure of output (AT) and an alternative measure of technology (TE) raises questions regarding model specification. In this rejoinder, we discuss these two areas of concern in order to address any previous omissions, needed clarifications, or differences in purposes. Because different methods were used in the two investigations, a third issue concerning methodologies seems relevant. Therefore we also include some general comments regarding "best-practice"and "aver-practice"methods of analysis.

167. Hjalmarsson, L.; Veiderpass, A. Kluwer Academic Publishers: Dordrecht, 1992.
This article examines the efficiency of retail distributors in Sweden by use of different versions of the non-parametric Data Envelopment Analysis (DEA) method. Comparisons are made between different types of ownership and between different types of service areas.

The study indicates a rather low level of technical efficiency, a high level of scale efficiency in urban areas, but a fairly low level of scale efficiency in rural areas. The results show no significant differences in efficiency between different types of ownership or economic organization.

168. Hotes, R. W. Combined Proceedings: Sixth Annual Conference on Interactive Instruction Delivery and Third Annual Conference on Learning Technology in the Health Care Sciences ; pp 78-84.
Data envelopment analysis (DEA) is a statistical technique which allows comparison of service delivery units having differing inputs and outputs. Although DEA techniques have not been widely used for training efficiency analysis, the methodology, based on linear programming procedures, has been demonstrated to be valid in situations where decision-making units with varied inputs and outputs must be evaluated. The author describes an application of a DEA-based evaluation software algorithm to an international training organization. Special consideration is given to evaluating instructional delivery systems using interactive video.
Computer aided instruction. Interactive systems. Linear programming. Performance evaluation. Statistical analysis. Video and audio discs.

169. Howard, L. H.; Miller, J. L. Academy of Management Journal 1993, 36(4), 882-94.
Data envelopment analysis shows promise for generating objective estimates of pay equity. Applied to performance and salary data for 433 professional baseball players, data envelopment analysis identified objectively underpaid, equitably paid, and overpaid players. The output also identified relevant referents for each player and quantified the relative influences of referents and all variables.

170. Huang, Y.-G. L. Journal of Medical Systems 1989 , 13(3), 155-162.

171. Huang, Y.-G. L.; McLaughlin, C. P. HSR: Health Services Research 1989, 24(2), 143-158.

172. Jagannathan, R. Management Science 1985, 31(7), 847-851.

173. Jesson, D. National Foundation for Education Research: 1988.

174. Jesson, D.; Mayston, D. Newbury, Berkshire, U.K., 1989.

175. Jesson, D.; Mayston, D.; Smith, P. Oxford Review of Education 1987, 13(3), 249-266.

176. Joseph, D. A.; Cerveny, R. P.; Rhodes, E. L. Proceedings of the AIDS National Meeting.

177. Kamakura, W. A. Management Science 1988, 34( 10), 1273-6.
A variant of Banker and Morey's (1986) DEA model for controllable ordinal outputs is proposed. As opposed to the original model, this version allows the comparison of a decision making unit (DMU) to other DMUs operating at equal or higher levels of the ordinal outputs.
Decision theory. Management science.

178. Kao, C. European Journal of Operational Research 1994, 73(3), 487-494.
Data envelopment analysis (DEA) is a mathematical approach to assessing relative efficiency within a group of decision making units (DMUs). An important outcome of such an analysis is a set of values for slack variables. Theoretically, the slack variables indicate by how much the associated factors (inputs or outputs) should be adjusted so that all inefficiencies will be eliminated. Nevertheless, in real world applications this is impractical because not all factors can be controlled at will. In this paper, a modified formulation of the DEA model is presented where bounds are imposed on each factor. Simple relationships between factors are also incorporated. In essence the results solved from the proposed model provide the top management with a direction which is feasible in reality for achieving Pareto efficiency.
management science, mathematical programming.

179. Kao, C.; Yang, Y. C. European Journal of Operational Research 1992, 72( 1), 43-51.
Junior colleges of technology (JCTs) in Taiwan are evaluated every three years by the government under five categories: eduational goal, instructors, curriculum, equipment, and administration. A lot of effort and expenditures are devoted to each evaluation. In this paper, a quantitative method based on the concept of Pareto-optimal organization which is much easier to conduct is proposed. This method is a simpler version of the data envelopment analysis (DEA) technique for measuring relative efficiency. Data from departments of industrial engineering and management of eleven JCTs were collected to illustrate how this method is applied to real world evaluation. Results from this quantitative evaluation coincide perfectly with that of the government evaluation. From a dual analysis, suggestions for improvement for each JCT are also derived.

180. Kao, C.; Yang, Y. C. European Journal of Operational Research 1992, 58( 3), 356-362.
The data envelopment analysis (DEA) approach is applied to find the inefficient forest districts in Taiwan ROC. The Taiwan Forestry Bureau proposes three alternatives for reorganizing the thirteen districts into eight districts. These three alternatives are appraised also by the DEA approach. The criterion is to make the new districts in an even status to be fair for the welfare of the local people. From the theoretical point of view, a better alternative for reorganization is suggested.
forestry. linear programming. operations research.

181. Karkazis J.; Boffey B. Working Paper 1995.
The paper addresses issues related with the spatial organisation of an industrial area using environmental criteria. Issues regarding environmental pollution, equity and efficiency are considered.

182. Kemerer, C. F. Proceedings of the Conference on Software Maintenance - 1988 (IEEE Cat. No.88CH2615-3) ; p 282.
Reports on efforts to develop and estimate a preliminary model of the software production process using pilot data from 65 software maintenance projects recently completed by a large regional bank's data processing department. The goals are to measure factors that affect software maintenance productivity, to integrate the quality and productivity dimensions of software measurements, and to examine the productivity of entire projects rather than only the programming phase, which typically accounts for less than half the effort on a software project. Variables relating to the quality of labor employed on the projects are included. To investigate the set of potential productivity factors, the technique of data envelopment analysis (DEA) is used to estimate the relationship between the inputs and products of software maintenance. The general approach to this research is to model software development as a microeconomic production process utilizing inputs and producing products.
DP management. Software engineering.

183. Kerstens, K.; Vanden Eeckaut, P. CORE Discussion Paper 9513 .

184. Kornbluth, J. S. H. Journal of the Operational Research Society 1991, 42( 12), 1097-104.
Reducing weight flexibility has been suggested as a method for ensuring that the solution to data envelopment analyses do not give unreasonably low weightings to certain inputs or outputs. The author extends the use of reducing weight flexibility and uses it to model the effects of the decision-making unit's (DMUs) objectives on its efficiency relative to other DMUs with possibly different objectives. It is shown how such an approach can identify situations in which the weights imputed by a data envelopment analysis can be inconsistent with the decision-making weights used by the firm, and how this approach can be used to provide efficiency measures that are consistent with the DMU's own objectives. The method allows the analyst to distinguish between a decision-making unit's technological inefficiency and its inability to implement its own policies.
Management science. Mathematical programming.

185. Korostolev, A. P.; Simar, L.; Tsybakov, A. B. 9405; On Estimation of Monotone and Convex Boundaries Institut de Statistitique: 1994.
We study the problem of estimating a set G in Rk (or equivalently, estimating its bounardy) given n independent identically distributed in G observations X1,...,Xn. We suppose that the boundary of G can be represented as a monotone or convex function of k-1 arguments. We evaluate the risks of several estimators of boundaries and show that they converge with the best possible rates. A general density of Xi's with the support G is considered, as well as the extension to the case where a small portion of "data outliers"falls out of the set G.
monotone boundary, convex boundary, free disposal hull, data envelopment analysis, estimation of support of a density, minimax risk.

186. Kumbhakar S.; Hjalmarsson L. Journal of Applied Econometrics 1995, 10, 33-47.
The paper uses panel data models and proposes a multi-stage approach for obtaining: parameters of an input requirement flexible function; technical efficiency decomposed into time-invariant (firm fixed), time varying and random noise. The method is applied on social security offices and Sweden.

187. Learner, D. B.; Phillips, F.; Rousseau, J. J. Proceedings of the 41st ESOMAR Congress on Marketing Research.

188. Levitt, M. S.; Joyce, M. A. S. Cambridge University Press: Cambridge, 1987.

189. Lewin, A. Y. Health Services Research 1983, 18(2), 206-208.

190. Lewin, A. Y.; Minton, J. W. Management Science 1986, 32(5), 514-538.

191. Lewin, A. Y.; Morey, R. C. International Journal of Policy Analysis and Information Systems 1981, 5 , 267-285.

192. Lewin, A. Y.; Morey, R. C.; Cook, T. J. Omega 1982 , 10(4), 401-411.

193. Lovell, C. A. L.; Schmidt, P. Kluwer Academic Publishers: Boston, 1988.

194. Lynde C.; Richmond J. Georgia Productivity Workshop 1994.

195. Macmillan, W. D. Papers of Regional Science Association 1986, 60, 41-57.

196. Macmillan, W. D. Environmental and Planning A 1987 , 19, 1511-1524.

197. Mahmood, M. A. Information Systems Journal 1994 , 4(2), 93-115.
Even during this recessionary era, IT expenditure in most organizations continues to grow at a high rate. Because of this increased IT expenditure, more and more senior managers are demanding that IT play a greater role in determining their firm's success by helping the increase organizational efficiencies and perhaps even achieve competitive advantage. The existing information systems literature on IT investment and organizational strategic performance, however provides very little help to senior managers in making before-the-fact IT investment decisions. This research study puts forth some 'hard' evidence relating IT investment to organizational strategic and economic performance by using data envelopment analysis (DEA). Eight IT investment measures were used as inputs and 10 organizational strategic and economic performance ratios were used as outputs for the DEA model. The results indicate that two-thirds of the organizations in this research study are deemed efficient by DEA. A clear distinction exists in terms of IT investment and organizational strategic and economic performance. The firms in the efficient group had a much higher return on their information technology investment than the inefficient group. In addition the DEA results pinpoint the inefficient inputs and deficient outputs for an inefficient firm allowing a senior manager to take corrective actions to compensate for the situation.
DP management, economic and sociologic effects, information systems, investment.

198. Maindiratta, A. Journal of Econometrics 1990 , 46, 57-72.
This paper extends the usual DEA analysis, which evaluates savings that could have been effected by a decision making unit, given its observed task, to inquire into whether even greateer savings would be possible if the task were to be optimally apportioned to a number of smaller unites. The notion of size efficiency is introduced to measure this potential for further input reductions, and then compared and contrasted to the eextant notion of scale efficiency. The existence of a largest radially size efficient output scale is established as a ray property of the production frontier. An illustrative application to hospitals is described.

199. Mayston, D.; Jesson, D. Oxford Review of Education 1988, 14(3), 321-339.

200. McDonald, J.; Moffitt, R. The Review of Economics and Statistics 1979, 318-320.
A discussion is given concerning the use and application of Tobit models.

201. Mensah, Y. M.; Li, S.-H. The Accounting Review 1993, 68(1), 66-88.
In a recent study, Hayes and Millar (1990) presented empirical evidence on the cost function and apparent cost-optimizing behavior of the local managers of 33 county jails in Tennessee. Two main arguments were advanced: (1) line-item budgeting (LIB) was an ineffective control because the cost shares are assumed fixed in such budgetary settings, and (2) the budgetary control and performance evaluation process in not-for-profit settings could be improvded if the underlying cost function was estimated by using a flexible functional form such as the translog to gain knowledge of possible input substiturion and output transformation. This translog budget model was veiewed implicity as a superior alternative to the more traditional LIB approach.

The main objective of this study is to extend Hayes and Millar's idea of atraslog budget model by outlining a more complete budgetary system. It is shown that a frontier cost function generated from the ordinarly least squares (OLS) translog function can be used to identify four types of inefficiencies: the degree of technical, allocative, and scale inefficiency as well as institutional X-inefficiency (Liebenstein 1966). These four types of inefficiencies are then linked to both the long-term and short-term objectives of budgeting; namely the performance evaluation, subordinate manager motivation, planning, and control.

The second objective of the study is to compare the translog budget model (as revised) against the most frequently applied alternative technique, the nonparametric data envelopment analysis (DEA), in the same context. From the standpoint of a routine budgetary control system, it was found that the DEA model is more suitable than the translog model. In not-for-profit settings, however, an econometric model such as the translog may be needed initially in the specification of the most appropriate input and output measures. Although the effectiveness of LIB as a control is not addressed directly, our analysis shows that the line-item data mandatory with this approach is an essential part of the effective budgetary system designed for a not-for-profit setting.
data envelopment analysis, line-item budgeting, production efficiency evaluation, x-efficiency.

202. Metzger, L. Journal of Applied Business Research 1992, 9(3), 69-79.
Quality has become more important than ever in the production of goods and services. Money spent on quality must ultimately contribute to increase productivity within the firm. Quantifying the benefits of quality costs with respect to increased productivity is an elusive proposition. This paper describes the use of a linear programming method, DEA, to measure the effects of apprisal and prevention costs on productivity.

203. Miliotis, P. A. Journal of the Operational Research Society 1992, 43( 5), 549-55.
Data envelopment analysis (DEA) is used to evaluate efficiency measures for the 45 distribution districts of the Greek Public Power Corporation (PPC). Results are derived under different sets of assumptions and are compared with simple productivity indices used by PPC and with efficiency measures produced by econometric methods. DEA scores appear to be more reliable than simple productivity indices. Comparison of the different cases explains the reason for the low efficiencies, which can be due to the management of controllable inputs, the design of the supply system or other environmental factors.
economics. linear programming. load distribution. management.

204. Miller, G. Public Productivity and Management Review 1990, XIII(4), 333-351.

205. Morey, R. C.; Capettini, R.; Dittman, D. A. Healh Services Research 1985, 18(2), 50-58.

206. Norman, M.; Stoker, B. John Wiley and Sons Ltd.: 1991.
To explore the application of data envelopment analysis to assessing economic performance. Numberous case studies are provided.

207. Nunamaker, T. Health Services Research 1983, 18(2), 183-205.

208. Nunamaker, T. Managerial and Decision Economics 1985, 6(1), 50-58.

209. Nunamaker, T. Managerial and Decision Economics 1988, 9(3), 255-256.

210. O'Leary, D. E. H. M. H. Proceedings of the IASTED International Symposium Artificial Intelligence Application and Neural Networks - AINN '90 ; p 258.
The purpose of the paper is three-fold. First, an overall framework of those characteristics that have been found critical to the assessment process is developed, based on a review and extension of the literature. Second, a mathematical model is developed that is designed to model the process, in order to determine which development teams and approaches are the most successful. Data envelopment analysis is used to model the process. As a part of that model, the paper develops the relevant input and output variables to the process of developing an expert system. third, a statistical approach is used to determine the importance of particular variables.
DP management. expert systems. software engineering.

211. Olesen, O. B.; Petersen, N. C. 1994.
A model for efficiency evaluation based upon the theory of chance constrianed programming is developed. The model uses a piecewise linear envelopment of confidence regions for observed stochastic mulitple inputs multiple output combinations in the DEA tradition. The model allows for exogenous decommposition of the total variation in data for each DMU. By varying certain probability levels the model can provide estimates of the sensitivity of efficiency scores regarding an unknown amount of noise in data. An application of the method in an evaluation of the research activit8ies in economic departments at Danish Universities is presented.

212. Oral, M.; Kettani, O.; Yolalan, R. IIE Transactions 1992, 24( 5), 166-76.
Productivity analysis and its strategic implications are not only important for manufacturing sector but equally essential for other sectors as well. Branch banking sector is no exception in this sense and banks have to operate more efficiently and effectively in an increasingly competitive environment to sustain or improve their relative positions. This paper discusses the methodology of an empirical study that was employed in analyzing the operating productivities of a set of 44 bank branches of a major commercial bank offering relatively homogeneous products in a multi-market business environment. The methodology was based on the concepts and principles of data envelopment analysis (DEA). The results of the study have indicated that this kind of productivity analysis is not only complementary to traditionally used financial ratios but also is a useful bank management tool in reallocating resources between the branches in order to achieve higher efficiencies.
banking. resource allocation.

213. Oral, M.; Yolalan, R. European Journal of Operational Research 1990, 46( 3), 282-94.
Discusses the methodology of an empirical study that was employed to measure the operating efficiencies of a set of 20 bank branches of a major Turkish Commercial Bank offering relatively homogeneous products in a multi-market business environment. The methodology was based on the concepts and principles of data envelopment analysis. The results of the study have indicated that this kind of approach is not only complementary to traditionally used financial ratios but also a useful bank management tool in reallocating resources between the branches in order to achieve higher efficiencies. It has been also observed that the service-efficient bank branches were also the most profitable ones, suggesting the existence of a relationship between service efficiency and profitability.
Banking. Operations research.

214. Ozcan, Y. A.; Cotter, J. J. An Assessment of Efficiency of Area Agencies on Aging in Virginia Through Data Envelopment Analysis 1993Forthcoming.
This study investigate the performance of area agencies on aging using a current analytical technique, data envelopment analysis, DEA. Results demonstrate the extent of inefficiency and how efficiency of individual area agencies can be improved relative to their peer agencies based on analysis of resources available and service units produced. Evaluating by organizational type, size, and rural-urban differences, results indicate governmental agencies, large agencies, and urban agencies are more likely to be efficient.
efficiency, data envelopment analysis, area agencies on aging, evaluaton.

215. Parkan, C. Eng. Costs & Prod. Econ. (Netherlands) 1987, 12( 1-4), 237-42.
Discusses an application of data envelopment analysis to bank branches to identify any operational inefficiencies. Data related issues and implementation difficulties are also included.
banking.

216. Pavlopoulos P.; Kouzelis A. Applied Economics 1989 , 21, 285-293.
The authors use a translog cost function to assess the cost efficiency of bank branches. Interesting feature is that they include the market structure and competition as input factors in the analysis.

217. Petersen, N. C. Management Science 1990, 36( 3), 305-14.
Data envelopment analysis (DEA) is a procedure for evaluation of the relative efficiency of units in a production system. DEA involves construction of composite units as convex combinations of other units' inputs and outputs under various conditions related to returns to scale. The assumption of convexity is inconsistent with increasing returns to scale and not implied by constant or non-increasing returns to scale. An alternative DEA approach which does not invoke the assumption of a convex production possibility set is developed in the paper. It is implemented in two stages. Stage one is concerned with the spanning of isoquants and stage two with an evaluation of the position of an observed input-output combination compared to the isoquants. Both stages involve linear programming techniques solely.
Linear programming. Management science. Probability.

218. Pettypool, M. D.; Troutt, M. D. Math. Comput. Model. (UK); pp 1101-3
429-44). Based on this work, a three step procedure is presented which may be used to estimate the values of parameters such as unknown costs. A method called the principle of maximum efficiency estimation is introduced. An application to the estimation of inventory cost parameters is given.
In 1978 Charnes, Cooper, and Rhodes introduced data envelopment analysis as a method to rate the efficiencies of decision-making units by imputing weights to observed inputs and outputs in order to form an efficiency ratio. (Eur. J. Oper. Res.
Decision theory. Mathematical programming. Maximum principle. Parameter estimation.

219. Pettypool, M. D.; Troutt, M. D. Math. Comput. Model. (UK); pp 1104-6
429-44, 1978) is examined briefly. Two important shortcomings of the model are addressed and then two new formulations are presented. An approach based on extended pivoting in the original formulation is given.
The original Charnes, Cooper and Rhodes formulation of data envelopment analysis (Eur. J. Oper. Res..
Decision theory. Mathematical programming.

220. Phillips, F.; Parsons, R. G.; Donoho, A. Computers, Environment and Urban Systems 1990, 14( 2), 167-70.
Data envelopment analysis (DEA) is an eminently parallelizable process. The multiple linear programming (LP) problems that must be solved in a DEA application need not be solved in any particular order and each LP can fit within a single 'transputer'. An experiment using DEA on a newly available Levco Translink (2.5 MFLOPS) board for the Macintosh personal computer is reported.
Apple computers. Linear programming. Microcomputer applications. Operations research. Parallel machines. Parallel programming.

221. Rangan, N.; Grabowski, R.; Aly, H. Y.; Pasurka, C. Economics Letters 1988, 28, 169-175.

222. Rao, R. Journal of Economics 1992, 55 (1), 17-41.
An optimal control framework is used to examine principal-agent issues emphasising equity issues. It was found that coaperative solutions cannot be sustained between management and workers even under conditions of perfect information. Also, the incetive scheme can only endure the cooperative solutions after they have obtained.

223. Ray, S. C. Socio-Economic Planning Sciences 1988 , 22(4), 167-176.

224. Ray, S. C. Management Science 1991, 37 ( 12), 1620-8.
Combines data envelopment analysis (DEA) with regression modelling to estimate relative efficiency in the public school districts of Connecticut. Factors affecting achievements are classified as school inputs and other socio-economic factors. DEA is performed with the school inputs only. Efficiency measures obtained from DEA are subsequently related to the socio-economic factors in a regression model with a one-sided disturbance term. The findings suggest that while productivity of school inputs varies considerably across districts that can be ascribed to a large extent to differences in the socio-economic background of the communities served. Variation in managerial efficiency is much less than what is implied by the DEA results.
Economic and sociologic effects. Education. Management science. Statistical analysis.

225. Rhodes, E. L. New Directions for Program Evaluation; American Evaluation Association, Jossey Bass, Inc.: San Francisco, 1986; Vol. 32.

226. Roll, Y.; Cook, W.; Golany, B. IIE Transactions 1991, 23( 1), 2-9.
Data envelopment analysis (DEA) is a mathematical programming approach to assessing relative efficiencies within a group of decision making units (DMUs). An important outcome of such an analysis is a set of virtual multipliers or weights accorded to each (input or output) factor taken into account. These sets of weights are, typically, different for each of the participating DMUs. A version of the DEA model is offered where bounds are imposed on weights, thus reducing the variation in the importance accorded to the same factor by the various DMUs. Techniques for locating appropriate bounds are suggested and the notion of a common set of weights is examined. Possible interpretations to differences in efficiency ratings obtained with the various models developed are discussed.
Management science. Mathematical programming.

227. Roll, Y.; Golany, B. Omega 1993, 21( 1), 99-109.
Provisions for controlling factor weights constitute a significant extension of the data envelopment analysis (DEA) methodology, as an effective tool for measuring efficiency. The paper suggests a conceptual framework for the treatment of factor weights in DEA. First, it proposes general guidelines for setting bounds on factor weights. Then, it develops and presents alternative methods to limit the range within which these factor weights are allowed to vary. All of these methods involve additional information which is entered into the analysis in the form of constraints, bounds or different objective functions. Finally, the implications of the various approaches is discussed.
DEA, efficiency, multidimensional scaling.

228. Roll, Y.; Golany, B.; Seroussy, D. European Journal of Operational Research 1989, 43( 2), 136-42.
Presents an application of data envelopment analysis (DEA) methodology to maintenance units in the Israeli Air Force. Some aspects of using the technique for measuring efficiency within a complex public organization are demonstrated and discussed. Emphasis is based on the choice of factors to enter the analysis and on assigning numerical values to qualitative factors. Relative efficiency ratings, using different reference sets, are used to construct a hierarchical efficiency monitoring system, by which performance of various levels of the organization is evaluated.
Maintenance engineering. Military systems. Operations research.

229. Rousseau, J. J.; Semple, J. H. Management Science 1993, 39( 3), 384-6.
A new linear programming formulation for handling categorical outputs in data envelopment analysis (DEA) is presented which eliminates the difficulties of interpretation and computation that accompanied earlier mixed integer models.
linear programming. management science.

230. Schary, P.; Kleinsorge, I. K.; Tanner, R. International Journal of Physical Distribution and Materials Management forthcoming.

231. Schefczyk, M. International Journal of Production Economics 1993, 32( 1), 1, 2-11.
The purpose of the paper is to review two techniques to analyze the performance of industrial entities. Performance analysis can be applied to benchmarking of facilities. A case study of industry data is the basis for the application of measures. The reference data set contains information on sixteen warehouses belonging to the nationwide service parts distribution network of a large manufacturer. First, the author documents the application of productivity ratios to warehouses. "data envelopment analysis"is the second technique. The results of both approaches are compared using the Spearman coefficient of rank correlation and linear regression analysis. The author concludes that simple cost-based ratios appear to be the most suitable for situations where pricing structures for resources and services are uniform and known, such as for company internal benchmarking.
corporate modelling. nonparametric statistics. operations research. statistical anal.

232. Schefczyk, M. Strategic Management Journal 1993 , 14, 301-317.
This study presents a new approach for measuring operational performance, an important facet of performance missing in the current literature concerned with interational airline strategy. International perfomance assessments of airlines from published financial information are difficult, because (1) most airlines lease a substantial fraction of their aircraft, and (2) different accounting and taxation rules in various contries result in different impacts of leased assets on profit and balance-sheet information. A possible soultion are nonfinancial data. For example, the number of available ton kilometers may reflect aircraft capacticy more accurately than flight equipment depreciation. Howeever, different units of measurement introduce new difficulties. Drawing on data from 15 airlines, this study utilizes 'Data Envelopment Analysis' as a technique to analyze and compare operational performance of airlines. The study concludes with an analysis of strategic factors of high profitability and performance in the airline industry.
data envelopment analysis, benchmarking, ariline performance, airline strategy, production economics.

233. Schinnar, A. P. Institute of Social Studies: Hague, Netherlands, 1980.

234. Schmidt, F. University of Georgia, Department of Management Scienc and Information Technology, 1990.

235. Seaver B.; Triantis K. Management Science (Forthcoming) 1995.

236. Seiford, L. M. Computers, Environment and Urban Systems 1990, 14( 2), 171-5.
A brief survey of data envelopment analysis (DEA) is provided, in an attempt to bring readers abreast of recent progress in DEA research and uses. Selected references are presented which cover the various DEA models, extensions to the methodology, and example applications in the areas of education, health care, regional planning and information systems.
Administrative data processing. Linear programming. Operations research.

237. Seiford, L. M.; Thrall, R. M. Journal of Econometrics 1990, 46, 7-38.
This paper discusses the mathematical programming approach to frontier estimation known as Data Envelopment Analysis (DEA). We examine the effect of model orientation on the efficient frontier and the effect of convexity requirements on returns to scale. Transformations between models are provided. Methodological extesnions and alternate models that have been proposed are reviewed and the advantages and limitations of a DEA approach are presented.

238. Semple, J.; Rousseau, J. J.; Haag, S. Forthcoming.
Sensitivity and robustness of efficiency classifications in the BCC and additive models of data envelopment analysis are addressed. The minimum distance (measured by a generalized Tchebycheff norm) separating an organization from reclassification is computed. This value is show to constitute a generalized "residual"for each organization. Analysis of these residuals within two groups from an earlier DEA study reveals differences in efficiencies which were heretofore undetectable.

239. Sengupta, J. K. Comput. & Oper. Res. (GB) 1987, 14( 2), 117-29.
The data envelopment analysis for measuring efficiency of decision-making units is generalized for stochastic variations of input and output data. The case of one output and many inputs is considered for three types of data variations, e.g. in the objective function, in the constraints and the outputs. It is shown that a minimax method of efficiency measurement through chance-constrained programming methods can be suitably applied for the case of chance constraints. Some empirical applications to measurement of efficiency of public schools are also analyzed to test the sensitivity and robustness of the efficiency ranking and measurement.
decision theory. management science. minimax techniques. operations research.

240. Sengupta, J. K. International Journal of Systems Science 1989, 20( 11), 2085-93.
Maximizing the correlation between inputs and outputs and outputs in data envelopment analysis is analysed in three respects: the minimization of the L/sub p/-norm; the standard regression approach; and the case of composite outputs. The implications of canonical correlation are explored and their applications to efficiency studies of public sector enterprises discussed. This shows that the correlation measure may involve in some sense a basic and fundamental extension of the efficiency model analysed by data envelopment analysis.
Correlation methods. Matrix algebra. Minimisation. Statistical analysis.

241. Sengupta, J. K. Kluwer Academic Publishers: 1989.

242. Sengupta, J. K. International Journal of Systems Science 1988, 19(7), 1085-1094.

243. Sengupta, J. K. Int. J. Syst. Sci. (UK) 1987 , 18( 12), 2279-304.
The new technique of data envelopment analysis is generalized for stochastic and dynamic cases with some empirical applications emphasizing issues of robustness and sensitivity.
Economic cybernetics.

244. Sengupta, J. K. International Journal of Systems Science 1982, 13(3), 273-287.
Measuring efficiency of systems with multiple inputs and multiple outputs is most difficult when market prices for these inputs and outputs are unavailable as for example in most public sector decision-making units. Two types of measures related to Pareto-efficiency could still be defined in such a framework, one related to the productivity of a common set of inputs and the other to a ratio of weighted index of outputs and inputs. These measures however are ordinal and useful in ranking purposes only and may have stochastic variations when the multivariate input-output system contains random components. Some characterizations and theoretical results are derived here for some specific types of random variations. Potential applications are also briefly indicated.

245. Sengupta, J. K. Computers & Mathematics With Applications 1992, 24( 8-9), 259-66.
The use of fuzzy set-theoretic measures is explored in the context of data envelopment analysis, which utilizes a nonparametric approach to measure efficiency. Three types of fuzzy statics are employed e.g., fuzzy mathematical programme, fuzzy regression and fuzzy entropy, to illustrate the types of decisions and solutions that are achievable, when the data are vague and prior information is inexact and imprecise.
data analysis. fuzzy set theory. management science.

246. Sengupta, J. K. Mathematical Programming, Series B 1991, 52( 1), 147-66.
The behavior of efficiency in the stochastic DEA model is examined in terms of the influence curve approach which quantifies the influence of observed data on the empirical fit of the production frontier. Data influence is analyzed through robustness. Two types of robustness and sensitivity issues are analyzed in terms of: (a) a new class of minimax measures; and (b) a set nonlinear efficiency measures and it is shown through empirical applications that in suitable cases these measures outperform the conventional ones.
Minimax techniques. Operations research.

247. Sengupta, J. K. Kluwer Academic Publishers: Dordrecht, 1985.

248. Sengupta, J. K. Managing and Decision Economics 1988, 9, 153-161.

249. Sengupta, J. K. International Journal of Systems Science 1989, 20( 2), 203-13.
Efficiency comparisons of two or more clusters under alternative information structures are attempted using a new interpretation of the data envelopment analysis model. Mixed strategies and information theoretic distance measures are used for the efficiency comparisons.
Information theory. Optimisation.

250. Sengupta, J. K. Fuzzy Sets and Systems 1992, 46( 1), 73-80.
Methods of measuring economic efficiency of input-output systems by employing a fuzzy statistical approach are explored in the context of the recent technique of data envelopment analysis, which utilizes a nonparametric approach to measure efficiency. Three applied fuzzy measures are illustrated: fuzzy linear programs: fuzzy linear regressions; and the fuzzy entropy measures in the context of a two-person game theory model.
Fuzzy set theory. Game theory. Linear programming. Statistical analysis.

251. Sengupta, J. K. International Journal of Systems Science 1993, 24( 11), 2159-73.
Previous methods of data envelopment analysis (DEA) deal mostly with static input-output systems. For dynamic systems with both current and capital inputs, the measurement of efficiency poses new problems due to intemporal effects. Some control-theoretic methods are discussed here by way of generalizing the static framework of data envelopment analysis in terms of dynamic multipliers and stochastic parameters. The dual problems of control and state estimation are analysed in their dynamic aspects, which show the myopic nature of measuring efficiency in a static version of the DEA model.
corporate modelling. decision theory. discrete time systems. linear programming. optimal control. state estimation.

252. Sengupta, J. K. Management Review 1986, 1, 3-18.

253. Sengupta, J. K. Kluwer Academic Publishers: Dordrecht, 1985.

254. Sengupta, J. K. Computers & Operations Research 1989, 16( 1), 55-65.
Some nonlinear methods of measuring efficiency in data envelopment analysis (DEA) are developed and compared with the linear measures. It is shown that suitable nonlinear measures may outperform the linear estimates in terms of goodness of fit and robustness. Hence they are more useful in extending the DEA model to industry-wide planning and other control environments.
Economics. Management science. Nonlinear programming. Operations research.

255. Sengupta, J. K. International Journal of Systems Science 1993, 24( 5), 857-71.
A class of nonparametric methods based on the minimax solution is developed for models of stochastic linear programming. These methods provide a measure of robustness through the adoption of a cautious policy. The usefulness of these methods is illustrated through data envelopment analysis which utilizes an optimizing method of efficiency measurement.
linear programming. minimax techniques. nonparametric statistics. stochastic program.

256. Sengupta, J. K. Managerial and Decision Economics 1987, 8(2), 93-99.

257. Sengupta, J. K. Kluwer Academic Publishers: Dordrecht, 1988.

258. Sengupta, J. K. Kluwer Academic Publishers: Dordrecht, 1985.

259. Sengupta, J. K. Computers & Operations Research 1991, 18( 2), 221-32.
Develops two broad measures of robustness in data envelopment analysis for efficiency measurement. One measure characterizes the probability distribution of random states of nature which generate the observed input-output data. On using this distribution, the sensitivity of the efficiency parameters can be evaluated. The second measure introduces a minimax class of optimal solutions, which have their robustness properties in a game-theoretic sense. These methods are developed by a set of theorems and empirically illustrated for situations of uncertainty, when only some partial information is available. These methods assume inadequate knowledge of the decision-maker about the random states of nature and develop a cautious optimal policy by testing if the chosen strategy or solution is very sensitive to the worst contingency that may arise. Some empirical applications to educational production functions show the usefulness of these operational methods.
Decision theory. Economic cybernetics. Education. Minimax techniques. Probability. Sensitivity analysis.

260. Sengupta, J. K. International Journal of Systems Science 1988, 19( 5), 779-91.
Two types of robust solutions of a class of efficiency models known as data envelopment analysis are explored. One is the minimax method and the other is related to the concept of a core in the theory of n-person games where the payoffs are random. Some potential applications of these robust solutions are also briefly discussed.
Game theory. Minimax techniques.

261. Sengupta, J. K. International Journal of Systems Science 1990, 21( 6), 1047-56.
Concepts of structural and industrial efficiency are developed for comparing productive efficiencies of two or more clusters and their mathematical characterizations worked out in the framework of data envelopment analysis. Some criteria of comparison of two or more similar clusters or aggregates of firms are developed and their relations with stochastic dominance and statistical distance between distributions explored.
Operations research. Statistical analysis.

262. Sengupta, J. K. Computers & Operations Research 1990, 17( 2), 123-32.
Canonical correlation theory is utilized to modify the set of efficiency measures in data envelopment analysis (DEA). This involves: (1) the mean square error criterion, (2) the canonical correlation test, and (3) the set of redundancy tests. An empirical application to educational production frontier data shows the usefulness of these tests and how they can be statistically tested as a two-stage procedure. This may help provide the bridge between the parametric and the nonparametric approaches to the estimation of production frontiers.
Decision theory. Operations research.

263. Sengupta, J. K.; Sfeir, R. E. Economics of Education Review 1986, 5(3), 297-307.

264. Sengupta, J. K. Kluwer Academic Publishers: Dordrecht, 1993; Vol. 25.

265. Sengupta, J. K. Journal of Econometrics 1990 , 46, 109-123.
A set of transformations based on the Lp-norm are used to generate alternative parametric and nonparametric formulations of production frontiers. Particular attention is paid to the cases p=1 (least absolute value estimation) and p=infinity (Chebyshev estimation). A class of generalized transformations of inputs and outputs by using the Box-Cox transformations is also explored and illustrated.

266. Sengupta, J. K.; Sfeir, R. E. Applied Economics 1988, 20, 285-293.

267. Sexton, T. R. New Directions for Program Evaluation; American Evaluation Association, Jossey Bass, Inc.: San Francisco, 1986; Vol. 32.

268. Sexton, T. R.; Silkman, R. H.; Hogan, A. New Directions for Program Evaluation; American Evaluation Association, Jossey Bass, Inc.: San Francisco, 1986; Vol. 32.

269. Sexton, T. R.; Sleeper, S.; Taggart, R. E. Jr. Interfaces 1994, 24(1), 87-103.
North Carolina uses data envelopment analysis (DEA) to produce a pupil transportation funding process that encourages operational efficiency and reduces expenditures. To do so, we extended the DEA methodology to nonhomogeneous units by integrating DEA with a regression model that adjusts DEA output to account for variations in site characteristics and to ensure that the final funding allocations were fiar. The new process has led to changes in bus routes and schedules, adjustments in school start and stop times, and reductions in the inventory of new buses. Between 1990 and 1993, the state saved $25.2 million in capital costs and $27.9 million in operating costs and it expects savings to increase.
education, resource allocation, statistical analysis, transportation.

270. Shakun, M. F.; Sudit, E. F. International Journal of General Systems 1983, 9(4), 205-215.

271. Sherman, H. D. Auditing - A Journal of Practice and Theory 1984, 4(1), 35-53.

272. Sherman, H. D. Medical Care 1984, 22(10), 922-938.

273. Sherman, H. D. Sloan Management Review 1984, 25(3), 11-23.

274. Sherman, H. D. New Directions for Program Evaluation; American Evaluation Association, Jossey Bass, Inc.: San Francisco, 1986; Vol. 32.

275. Sherman, H. D. The Society of Management Accountants of Canada: Hamilton, Ontario, 1988.

276. Sherman, H. D.; Gold, F. Journal of Banking and Finance (Netherlands) 1985, 9(2), 297-315.

277. Sickles, R. C.; Streitwieser, M. L. Kluwer Academic Publishers: Dordrecht, 1992.
Firm-specific and temporal patterns of efficiency of the interstate natural gas transmission industry during the implementation of the Natural Gas Policy Act are estimated by two alternative methodologies. A new panel stochastic frontier systems estimator exploits the potential exogeneity of certain regressors from firm effects. This allows for heterogeneity in slopes, as well as in intercepts. Patterns of technical efficiency based on the structural stochastic model are compared with those based on deterministic programming methods, date envelopment analysis. Concordant findings based on these alternative methodologies suggest a perversive pattern of declining efficiency in the industry during the period of phased in well-head price deregulation.

278. Silkman, R. H. New Directions for Program Evaluation; American Evaluation Association, Jossey Bass, Inc.: San Francisco, 1986; Vol. 32.

279. Silkman R.; Young D. National Tax Journal 1982 , 35(2).
Data from public libraries and School bus transportation are used to investigate the presence of technical inefficiency in th operation of local authorities. Econometric modesl are employed for this assessment.

280. Sinuany-Stern, Z.; Mehrez, A.; Barboy, A. Computers and Operations Research 1994, 21(5), 543-546.
The authors present a case study where academic departments at Ben-Gurion University, Israel, were evaluated via data envelopment analysis (DEA) using the CCR model by A. Charnes, W. W. Cooper, and E. Rhodes (1978). Extensive post analalyses were performed in several directions. First, various sets of data were used to identify efficient and inefficient departments. New efficiency measures are suggested in relation to the reference set included in the analyses of academic departments. We measured the efficiency of departments within the same school. We applied cluster analyses to divide the departments into several sets; and the discriminant analysis to test the match of the efficiency/inefficiency division of the CCR ratio. We further tested organizational changes where an inefficient department was closed and joins other departments. Finally we compared the CCR model to the pure economic approach-the cost per student ratio.
decision theory, education, nonlinear programming, operations research.

281. Smith, P. Omega 1990, 18( 2), 131-8.
Ratio analysis has been a tool of analysts for as long as financial statements have been prepared. Yet its limitation to considering only one numerator and one denominator severely limits its usefulness. The paper extends the traditional ratio analysis to permit the incorporation of any number of dimensions of performance, using data envelopment analysis. The method produces measures of corporate efficiency, together with a wealth of supporting information. The strengths and weaknesses of the method applied to financial statements are appraised.
Data analysis. Finance.

282. Smith, P.; Mayston, D. Omega (GB) 1987, 15( 3), 181-9.
One method of pursuing efficiency in the public sector has been the publication of performance indicators for individual agencies. It is often unclear how these indicators should be interpreted in isolation. The paper uses data envelopment analysis to show how the data underlying performance indicators can be used to generate a single measure of efficiency for an agency. The method systematically adjusts for differences in the environment that different agencies face. The potential limitations of the technique for the purpose of inter-agency comparison are then discussed.
data analysis. public administration.

283. Smith, P. 15; Misspecification Bias in Data Envelopment Analysis University of York: 1993.
The use of data envelopment analysis for estimating comparative efficiency has become widespread, and there has been considerable academic attention paid to the development of variants of the basic DEA model. However, one of the principal weaknesses of DEA is that - unlike statistically based methods - it yields no diagnostics to help the user to determine whether or not the chosen model is appropriate. In particular, the choice of inputs and outputs depends solely on the judgement of the user. The purpose of this paper is to examine the implications for efficiency scores of using a misspecified model. A simple production process is set up. Simulation models are then used to explore the effects of applying misspecified DEA models to this process. The phenomena investigated are: the omission of significant variables; the inclusion of the irrelevant variables; and the adoption of incorrect returns to scale assumptions. The robustness of the results is investigated in relation to sample size; variations in the number of inputs; correlation between inputs; and variations in the importance of inputs. The paper concludes that the dangers of misspecification are most serious when simple models are and sample sizes are small. In such circumstances, it is concluded that it will usually be to the modeller's advantage to err on the side of including possibly irrelevant variables rather than run the risk of excluding a potentially important variable from the model.
Sensitivity DEA misspecification.

284. Sojka, J. Ekonomicko-Matematicky Obzor 1989 , 25( 2), 198-207.
After the methodology of data envelopment analysis elaborated by A. Charnes and his colleagues emerged, the author started its analysis with a view to measure its effectiveness in socialist countries. Relevant research was realized in cooperation with the colleagues of the author's department, of research institutes, or with the help of students. In all the research work, this method has proved effective though not without problems. In the first part of the paper the author presents some problems which have been solved, in the second part, the author discusses some difficulties which have been met in this regard.
Decision theory. Management science.

285. Stewart, T. J. OMEGA International Journal of Management Science 1994, 22(2), 205-206.

286. Stolp, C. Computers, Environment and Urban Systems 1990, 14( 2), 103-16.
The popularity of data envelopment analysis (DEA) as a tool for examining the technical efficiency of 'decision making units' (DMUs) has spread widely and rapidly since its original introduction by E.L. Rhodes (1978). The author adopts a pragmatic perspective and-while neither damning DEA nor offering it as a panacea for productivity analysis-seeks to underscore some of its strengths as well as its weaknesses. While speculating on promising applications of DEA to urban and regional analysis in such disciplines as planning, geography and regional science, he reviews the basic logic of the method and highlights the key features it offers analysts interested in examining the relative technical efficiency of DMUs. Some statistical and information-theoretic concerns center on what appears to be DEA's underutilization of potentially useful empirical information. He also suggests how DEA can be viewed both in the light of more conventional statistical methods and as an informative and useful tool for systematic sensitivity analysis. Additional concerns and considerations in applying DEA are also outlined.
Decision theory. DP management. Linear programming. Operations research. Town and country planning.

287. Stolp, C. Westview Press: Boulder, Colorado, 1987.

288. Sueyoshi, T. Journal of the Operations Research Society of Japan 1992, 35( 1), 62-76.
An algorithmic strategy is proposed for use with the assurance region (AR) approach in data envelopment analysis (DEA). The strategy addressed characterizes and classifies all decision making units (DMUs) into several subsets, using the revised simplex method of linear programming. Then, each DMU subset is solved by a different algorithm. Experimental studies consisting of randomly generated data sets have confirmed that the proposed algorithm outperforms the conventional DEA use of the revised simplex method. An important feature related to the DEA/AR algorithm is that it can deal effectively with large data sets.
decision theory. linear programming.

289. Sueyoshi T. European Journal of Operational Research 1994, 77, 253-271.
The paper examines salary discrimination questions using goal programming formulations.

290. Sueyoshi, T. Journal of the Operational Research Society 1991, 42( 6), 463-77.
Develops a new use of data envelopment analysis for estimating a stochastic frontier cost function that is assumed to have two different error components: a one-sided disturbance (representing technical and allocative inefficiencies) and a two-sided disturbance (representing an observational error). The two error components are handled by data envelopment analysis in combination with goal programming/constrained regression. The approach proposed can avoid several statistical assumptions used in conventional methods for estimating a stochastic frontier function. As an application, this study uses the estimation technique to obtain an AT&T stochastic frontier cost function.
Economics. Estimation theory. Mathematical programming. Stochastic processes.

291. Sueyoshi, T. Journal of the Operational Research Society 1992, 43( 2), 141-55.
Presents an effectively-designed algorithm for measuring technical, allocative and overall efficiencies, using data envelopment analysis (DEA). The specially designed DEA code takes advantage of unique features related to the DEA algorithm. Computational efficiency of the proposed DEA algorithm is confirmed by a Monte Carlo simulation study.
Linear programming. Management science. Operations research.

292. Sueyoshi, T. Journal of the Operational Research Society 1990, 41( 3), 249-57.
A special algorithm is presented for the additive model in data envelopment analysis (DEA). The special algorithm first classifies a data set into several subsets. Then the subset is solved by a different algorithm framework. In simulation studies, the algorithm outperformed available DEA codes. The proposed algorithm can efficiently deal with a large data set.
Management science. Nonparametric statistics. Operations research. Set theory.

293. Sueyoshi, T.; Chang, Y.-L. Operations Research Letters 1989, 8( 4), 205-13.
An efficient algorithm is proposed for the additive and multiplicative models in data envelopment analysis (DEA). In simulation studies the algorithm executed in less than 60% of the CPU time required by the revised simplex method.
Decision theory. Linear programming.

294. Swann, G. M. P. Applied Economics 1987, 19, 201-213.

295. Thanassoulis, E. Journal of the Operational Research Society 1993, 44( 11), 1129-44.
This paper compares regression analysis and data envelopment analysis as two alternative methods for assessing the comparative performance of homogeneous organizational units such as bank branches or schools. The comparison is restricted to units using a single resource or securing a single output. It focuses on the estimates of relative efficiency, marginal input-output values and target input-output levels that the two methods offer. A set of hypothetical hospitals is used to illustrate the performance of the two methods. It is found that, in general, data envelopment analysis outperforms regression analysis on accuracy of estimates but regression analysis offers greater stability of accuracy.
health care. management science. operations research. optimisation. statistical anal.

296. Thanassoulis, E.; Dyson, R. G. European Journal of Operational Research 1992, 56( 1), 80-97.
Develops models which can be used to estimate alternative input-output target levels to render relatively inefficient organisational units efficient. The models can incorporate preferences over potential improvements to individual input output levels so that the resultant target levels reflect the user's preferences over alternative paths to efficiency. The paper illustrates the practical usefulness of the models developed and highlights the alternative measures of relative efficiency implicit in the models developed.
Decision theory. Operations research.

297. Thanassoulis, E.; Dyson, R. G.; Foster, M. J. J. Oper. Res. Soc. (GB) 1987, 38( 5), 397-411.
Examines the nature of information obtained from data envelopment analysis (DEA) in comparative studies of the efficiency of decision-making units, and discusses the interpretation and practical usefulness of such information. The themes developed in the paper are illustrated by an application of DEA to data on the rate-collection function of London Boroughs and Metropolitan District Councils. The paper begins with an overview of DEA, followed by a discussion of some of the practical considerations arising in the application of DEA. It then describes the structuring of the rate-collection function for assessment by DEA, and explores the extent to which units can be classified as relatively efficient or inefficient.
finance. management science.

298. Thomas, D.; Greffe, R.; Grant, K. Proceedings of the Fifth NARUC Biennial Regulatory Information Conference; The National Regulatory Research Institute, Ohio State University.

299. Thomas, H.; R. Frost Public Productivity and Management Review 1990, XIII(4), 369-385.

300. Thompson, R. G.; Singleton, F. D. Jr.; Thrall, R. M.; Smith, B. A. Interfaces (USA) 1986, 16( 6), 35-49.
In 1984, the Texas Legislature funded a four-university, inter-disciplinary effort to identify feasible sites for location of a very high-energy physics lab in Texas and to evaluate the comparative advantages of one site versus another. Six feasible sites were identified and a comparative site analysis was made by applying data envelopment analysis incorporating project cost, user time delay and environmental impact data. In addition, for the efficient sites, the price weights for user time delay and environmental impact, given normalization on project cost, were analyzed and arguments were developed to bracket these pairs of price weights into an 'assurance region' for the preferred site(s). the South Dallas site was found to be preferred for a wide range of conditions, while the North Houston site was sensitive to the method of indexing the impact of the environment. The method appears to be applicable to a wide range of siting problems faced by not only government, but also industry.
laboratories. operations research.

301. Thompson, R.; Lee, E.; Thrall, R. M. Computers & Operations Research 1992, 19( 5), 377-91.
The productive efficiencies of 45 randomly sampled oil/gas independent firms were analyzed year-by-year by application of data envelopment analysis (DEA) methods for the 7 years 1980-1986. This approach differs fundamentally from a 'window' analysis, where structural differences in the efficiency distributions over time may not be easily discerned. Both DEA ratio and convex models were applied; also, bounds were placed on the modeled prices (multipliers) by the use of assurance region (AR) principles. The distributions of DEA-efficiency measures were found to have significantly different characteristics in the 1980-1982 period than in the 1983-1986 period. Constant returns-to-scale prevailed in each of the 7 years. The AR-efficiencies in the presence of multiplier bounds, refined appreciably the candidate set of firms for overall efficiency. The AR-efficiency distributions exhibited significantly different levels and spreads in the 1980-1982 period than in the 1983-1986 period. Very few (5 at most) of the candidates for overall efficiency were viable economic firms in 1985 and 1986. Strikingly different energy policy conjectures follow from this micro analysis than from the US Department of Energy's macro analysis in its Energy Security study.
Energy resources. Natural gas technology. Oil technology. Petroleum industry.

302. Thompson, R.; Thrall, R. New Directions in Computational Economics 1994, 121-133.
The duality between efficiency and zero profit maxima has long characterised the theory of perfect competition in economics. DEA models in ratio and convex form imply zero profit maxima under the normalisations required for the linear programming reductions. Such zero profits are implied both in the absence and in the presence of cone-ratio asurance region bounds on the multipliers. However, if the AR input-output bounds on the multipliers are linked (linked cones) which is precluded with cone ratio then the normalisations required for the LP solution must be dispensed with and the DEA model need to reformulated to be meaningful. This LC reformulation shows efficiency and profitability are are separate concepts which gives new absolute profitability and non-linear efficiency measures. Both measures are relative to the full (m+s) dimensions of the multiplies spaces in contrast to the m+s-2 dimensions of the LP normalised multiplier spaces. For the multiple input/output problem computational procedures can be used to find the maximum profit solutions; for the one-output problem a non-linear programming procedure is suggested to find the efficiency solutions. Additional research is required to find the efficiency solutions for multiple output problems.

303. Thompson, R. G.; Dharmapala, P. S.; Thrall, R. M. Journal of Productivity Analysis 1993, 4, 379-390.
Some prominent literature in the DEA field has displayed the following three problems involving zeros: (1) The structural role of zeros in data has not necessarily been recognized; in fact, without explanation, arbitrary, small positive numbers have been substitute for such zeros. (2) It is not well recognized that the artificial non-Archimedean construct is not necessarily needed to exclude zero multipliers and to identify positive slacks. (3) Because of degeneracy, optimal solutions are not necessarily unique; this lack of uniqueness is especially important in interpretation of the multipliers and slacks.
DEA zeros.

304. Thompson, R. G.; Langemeier, L. N.; Lee, C.-T.; Lee, E.; Thrall, R. M. Journal of Econometrics 1990, 46 , 93-108.
As long recognized, the problem of efficiency invlolves both technical and economic facets. Determination of the technically efficienct firms provides the base for economic analysis. Values in terms of prices or costs must be introduced into the problem to work towards finding firms which might be regarded as overall efficient.

That problem came to the forefront in a 1984 study to find the best site for location of a Superconducting Super Collider (SSC) in the state of Texas. Application of a modern value-free frontier method called DEA to the data, which was primarily engineering and geological in character, showed five of the six feasible sites were technically efficient. However additional socioeconomic/environmental data provided 'price-cost' inequality bounds for the mathematical multipliers in the DEA problem. Including those bounds in the analysis, which was called an Assurance Region (AR), reduced the number of efficient sites from five to one. In 1988, the U.S. Department of Energy in national competition actually selected the site identified by the bounding method for location of the SSC.

In this paper, the AR concept is defined for efficiency analysis of the linear production possibility set. As applied here to 83 farms, we use only the special case of AR consisting of separate linear homogeneous restrictions on the input and ouput multipliers. When applied to the technically efficient farms, the AR principles reduced the number of candidates for overall efficiency from 23 to 8 in one case (ratio model) and from 44 to 13 in another case (convex model).

305. Thrall, R. M. Financial Accountability and Management 1989, 10, 159-162.

306. Tompkins, C.; Green, R. Financial Accountability and Management 1988, 4(2), 147-164.

307. Tone, K. International Journal on Policy and Information 1989, 13( 2), 57-64.
Both the analytic hierarchy process (AHP) and data envelopment analysis (DEA) aim at making decisions under multiple criteria environments. The AHP uses pairwise comparisons and eigenvector weightings, whereas DEA does linear fractional programmings. Some structural similarities between them are pointed out by a comparison of the benefit/cost analysis by AHP and DEA. Also, fixed and variable weights in multiple criteria decision-making are discussed.
Decision theory. Linear programming. Nonlinear programming.

308. Tulkens, H. Journal of Productivity Analysis 1993 , 4, 183-210.
The methodology of free disposal hull (FDH) measure of productive efficiency is defined and put in perspective vis-a-vis other nonparametric techniques, in terms of the postulates on which they respectively rest. Computational issues are also considered, in relation to the linear programming techniques used in DEA. The first application bears on a comparison between a private and a public bank, in terms of the relative efficiency of their branches. Important characteristics of the data are revealed by FDH that are not by DEA, due to a better data fit. Next, efficiency estimates of judicial activities are used to evaluate what part of the existing backlog could be reduced by efficiency increases. Finally, with monthly data of an urban transit firm over 12 years, the FDH methodology is extended to a sequential treatment of time series, that supplements efficiency estimation with a measure of technical progress.
fdh free disposal hull.

309. Valdmanis, V. Journal of Public Economics 1992 , 48, 185-205.
The research presented here applies the Data Envelopment Analysis (DEA) to a sample of public (government-owned) and not-for-profit hospitals operating in Michigan in 1982. The purpose of this research is twofold. First, assessing the relative efficiency between the these two ownership types using the DEA rather than cost or profit functions can add insights into the production practices of these two ownerships forms of hospitals. Second, testing the sensitivity of DEA will add credence to this approach and resulting measures. Ten different specifications of DEA are employed to test the sensitivity of the model.
DEA sensitivity hospitals.

310. Vassiloglou, M.; Giokas, D. Journal of the Operational Research Society 1990, 41( 7), 591-7.
Data envelopment analysis (DEA) has become an accepted approach for identifying inefficient decision-making units in an organization. The paper presents a systematic application of DEA carried out at the Commercial Bank of Greece in assessing the relative efficiency of bank branches. After a description of the model and the data, the results of the analysis are discussed, and a note is made of certain aspects of the follow-up analysis.
Banking. Decision theory. Linear programming.

311. Whittaker, G. Applied Economics 1994, 26, 469-479.

312. Wilson, P. Journal of Business and Economic Statistics 1993, 11, 319-323.
The articcle provides a statistical methodology for identifying outliers in production data with multiple inputs and outputs used in deterministic nonparametric frontier models. The methodology is useful in identifying observations that may contain some form of measurement error and thus merit closer scrutiny. when data checking is costly, the methodology may be used to rank observations in therms of their dissimilarity to other observations in the data, suggesting a priority for further inspection of the data.

313. Wong, Y.-H. B.; Beasley, J. E. Journal of the Operational Research Society 1990, 41( 9), 829-35.
Presents a method, based on the use of proportions, for restricting weight flexibility in data envelopment analysis. This method is applicable when the decision-making units being evaluated have multiple inputs and outputs.
Data analysis. Operations research.

314. Zhu J.; Shen Z. European Journal of Operational Research 1995, 81, 590-596.
The question of identification of returns to scale using DEA is addressed. The authors manage to defit the earlier paper by Chang and Chu and they defend the Banker definition and criterion for assessing economies of scale and also defining the concept of most productive scale size.

315. Athanassopoulos, A. and E., Thanassoulis, (1995), "Separating Efficiency from Profitability and its Implications for Planning", Journal of Operational Research Society, Vol. 46(1), pp. 30-45.

316. Athanassopoulos, A. and J., Storbeck, (1992), "Non-parametric models for assessing Spatial Efficiency", (Forthcoming in the Journal of Productivity Analysis).

316. Athanassopoulos, A. and N., Tatsos, Econometric Analysis as an Aid for Resource Allocation Decisions: The Case of Greek Local Authorities, Published Proceedings: Econometrics in Europe 2000, Th

317 Athanassopoulos, A. and Thanassoulis, E., (1995), Assessing marginal impacts of investments on organisational performance, Warwick Business School research papers No. 65 , International Journal

318. Athanassopoulos A., (1995), Performance Improvement Decision Aid Systems (PIDAS) in profit making organisations using data envelopment analysis", Journal of Productivity Analysis, Vol. 6 (2)

319. Athanassopoulos A. and J. Ballantine, (1995), "Ratio and frontier analysis for assessing corporate performance: Evidence from the grocery industry in the UK", Warwick Business School research p

320. Allen R., A.D. Athanassopoulos, R.G. Dyson & E. Thanassoulis, (1994), Weight restrictions & value judgments in data envelopment analysis: Evolution, development & future directions, Warwick Bus

321. Dyson R.G., E. Thanassoulis, and A. Athanassopoulos, (1993), Weight restrictions, production possibility sets and production functions, Warwick Business School research papers No.97.

322. Dyson R.G., A. Athanassopoulos and E. Thanassoulis, (1994), Performance measurement systems, Managerial control & data envelopment analysis, Warwick Business School research papers No. 117. (In

323. Athanassopoulos A. and R.G. Dyson (1994), Mathematical programming models for increasing flexibility in the assessment of the industry production function, Warwick Business School research pap

324. Athanassopoulos A. (1994), The evolution of non-parametric frontier analysis for assessing performance: Current practice and future developments. Forthcoming in Spoudai an International Journa

325. Athanassopoulos A. (1994), Flexible frontier analysis models for target setting and decision support: An application to electricity generating plants. Forthcomming in Decision Sciences.

326. Athanassopoulos A. (1994), Goal programming and Data Envelopment Analysis (GoDEA) models for multi-level multi-unit organisations: An application to Greek local authorities, Warwick Business S

327. Athanassopoulos A. (1994), Frontier analysis models for assessing comparative spatial disadvantage (CSD) of regions in the European Community, forthcomming in the European Journal of Operation

328. Athanassopoulos A. and O. Toivanen, (1995), "Production functions in multi-market oligopoly: Theory and evidence", Warwick Business School research paper.

329. Athanassopoulos A. and S. Curram, (1995), "A comparison of non-parametric methods as tools for assessing relative efficiency of decision making units", Warwick Business School research paper N

330. Athanassopoulos A., (1994), "Using technical efficiency as an aid to compare for-profit and not-for-profit organisations", Eds. Sakis Karagiorgas foundation, Athens, 1994.

331. Athanassopoulos A., (1995), "The usefulness of post-optimality indices to discriminate among relatively efficient units in data envelopment analysis", Warwick business school research paper.

332. Athanassopoulos A. and K. Triantis (1995), Assessing aggregate cost efficiency & policy implications of the Greek local authorities: a non-parametric approach, Warwick business scool paper.

333. Athanassopoulos A. and G. Karkazis (1995), Assessing the regional image effectiveness of prefectures of northern Greece using non-parametric frontier analysis, Warwick business school research

334. Athanassopoulos A. (1995), Multivariate and fronteir analysis for assessing the market and cost efficiency of large scale bank branch networks, Warwick Business School paper.

A. Emrouznejad © July, 1995. All rights reserved.