Araştırma Makalesi
BibTex RIS Kaynak Göster

An Automatic Input Analyzer for Simulation

Yıl 2003, Cilt: 2 Sayı: 3, 31 - 47, 15.12.2003

Öz

It is well known that the success of a simulation study depends on the success of the model of the real system. Based on this fact, in this study, it is aimed to develop a computer system for automatically diagnosing the theoretical distribution which models the data obtained from a stochastic system best. The system developed, named GAYUS, is intended for the naive users, and has the knowledge of several univariate theoretical distributions. In the decision making process, GAYUS uses a hybrid approach to integrate various heuristics, goodness-of-fit tests and graphical displays. Having these features, the system may be an example for the "simulation-based software”, or for the "cognizant front-end." A group of users tested the performance of the system. In the testing process, its users utilized random samples generated by the method of Monte Carlo. According to the users, GAYUS has an easy-to-use and a friendly interface. The test results indicated that as the sample size increases, the system better predicts the correct theoretical distribution from which the sample is drawn. We believe that the performance of GAYUS could be improved by increasing the variety of criteria used, and by utilizing more powerful input analysis techniques in the diagnosis process.

Kaynakça

  • BANKS, J. ( 1998a), Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice, New York: John Wiley and Sons.
  • BANKS, J. (1998b), Software for Simulation, J. Banks (Der), Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice İçinde, New York: John Wiley and Sons, 813-835.
  • CHAMBERS, J., CLEVELAND, W., KLEINER, B. ve TUKEY, P. (1983), Graphical Methods for Data Analysis, California: Wadsworth Int. Group.
  • D'AGOSTINO, R.B. ve STEPHENS. M.A. (1986), Goodness-of-Fit Techniques, New York: Marcel Dekker.
  • DANIEL, W. (1978), Applied Nonparametric Statistics, Boston: Houghton Mifflin Com.
  • ELIAS, M. (1996), Building Expert Systems: Principle Procedures and Applications, Minneapolis: Wadsworth .
  • FREUND, J.E. ( 1992), Mathematical Statistics. New Jersey: Prentice Hall.
  • GIARRATANO. J. ve RILEY. G. (1994). Expert Systems: Principles and Programming, Boston: PWS Pub.
  • GORDON, G. (1978), System Simulation New Jersey: Prentice Hall Inc .
  • GRAY, C. ve STRAIN, T. (1995), Visual Basic 4 Nuts and Bolts: for Experienced Programmers, California: McGraw Hill.
  • HAYES-ROTH, F., WATERMAN. A.D. ve LENAT, B.D. (1983), Building Expert Systems, New-York: Addison-Wesley.
  • HOAGLIN, D.C, MOSTELLER, F. ve TUKEY, J.W. (1983), Understanding Robust and Exploratory Data Analysis, Canada: John Wiley and Sons.
  • JOHNSON, N.L., KOTZ. S. ve KEMP, A.W. (1994), Continuous Univariate Distributions. Vol 1. Boston: Houghton Mifflin.
  • JOHNSON N.L., KOTZ. S. ve KEMP, AW. (1995). Continuous Univariate Distributions, Vol 2, Boston: Houghton Mifflin.
  • JOHNSON. N.L., KOTZ, S. ve KEMP. A.W. (1992), Univariate Discrete Distributions. Boston: Houghton Mifflin.
  • KELTON, W.D., SADOWSKl, R.P. ve SADOWSKI, D.A. (2002), Simulation with ARENA, Boston: McGraw-Hill.
  • KONVALINEN, J. ve WILEMAN. S. (1987). Programming with Pascal. Singapore: McGraw-Hill.
  • LAW. A.M. ve KELTON, W.D. (2000), Simulation Modeling and Analysis, New Jersey: McGraw-HiIl.
  • LAW. AM. ve McCOMAS, M.G. (1 999), ExpertFit: Total Support for Simulation Input Modeling, AP. Farrington. H.B. Newbhard. D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde, 261-266.
  • LEEMIS. L. (1999), Simulation Input Modelling, A.P. Farrington. H.B. Newbhard, D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde. 14-23.
  • MARIA, A. (1997), Introduction to Modeling and simulation, S. Andradottir, K.J. Healy, D.H. Withers ve B.L. Nelson (der.). Proceedings of Winter Simulation Conference içinde, 7-13.
  • NELSON, B.L. ve YAMNITSKY. M. (1998). Input Modelling Tools for Complex Problems. A.P. Farrington. H.B. Newbhard. D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde. 105-111.
  • NILSEN, N.R. (1991). Application of AI Techniques to Simulation, P.A. Fishwick, P.A ve R.B. Modjeski (der.), Knowledge-Based Simulation Methodology and Application içinde, New York: Springer-Verlag, 1-19.
  • ÖREN, T. (1994), Artificial lntelligence in Simulation, Annals of Operations Research. 63, 287- 319.
  • ÖZTÜRK, A. ve DUDEWICZ, E.J. (1992), a New Statistical Goodness-of-fit Test Based on Graphical Representation, Biometrica, 34, 403-427.
  • PROLOG DEVELOPMENT CENTER (1993), User's Book for Esta for Window, Denmark: PDC
  • RICH, E. (1983), Artificial Intelligence, New Jersey: MacGraw-Hill.
  • ROTHENBERG, J. (1990), Tutorial: Artificle Intelligence and Simulation, O. Balci, R.P. Sadowski ve R.E. Nance (der), Proceedings of Winter Simulation Conference içinde, 22-24.
  • RYAN. B.F., JOINER. B.L. ve RYAN, Jr. T.A. (1985), Minitab Handbook., Boston: PWS-Kent.
  • SCHMEISER, B. (1999) Advanced input Modeling for Simulation Experimentation, A.P. Farrington, H.B. Newbhard, D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulatian Conference içinde, 110- 115.
  • SELIA, A. (1995), Introduction to Simulation, C. Alexopoulos. K. Kang, W.R. Lilegdon ve D. Goldsman (der.), Proccedings of Winter Simulation Conference içinde. 7-14.
  • SHANNON, R.E. (1998), lntroduction to Art and Science of Simulation, DJ. Medeiros, E.F. Watson, J.S. Carson, M.S. Manivannan (der.), Proceedings of Winter Simulation Conference içinde, 7-14.
  • SHAPIRO, S.S. ve WILK, M.B. (1965), An Analysis of Variance Test for Normality, Biometrika, 52, 591-611.
  • TANIMOTO, S.L. ( 1995), The Elements of AI Using Common Lisp, Oxford: CS Press.
  • VINCENT, S. (1998), Input Data Analysis, J. Banks (der.), Handbook of Simulation: Principles, Methodology, Advances, Application, and Practice içinde, New York: John Wiley and Sons, 55-91.
  • YILMAZ, A. ve SABUNCUOĞLU, İ. (2000), lnput Data Analysis Using Neural Networks, Simulation,74(3),128-137.

Benzetimde Girdi Analizi Yapan Otomatik Bir Sistem

Yıl 2003, Cilt: 2 Sayı: 3, 31 - 47, 15.12.2003

Öz

Benzetim yönteminin başarısı, sistemin gerçeği yansıtacak şekilde modellenmesine bağlıdır. Bu nedenle bu çalışmada, rasgele bir sistemi en iyi modelleyen teorik dağılımı otomatik olarak saptayan bir bilgisayar sistemi geliştirmek amaçlanmıştır. Geliştirilen GAYUS isimli bu sistem, bilgi-tabanında tek değişkenli teorik dağılımlara ilişkin bilgiler içermektedir. Sistem karar verme sürecinde çeşitli sezgiler, iyi-uyum testleri ve grafik görüntülerden yararlanmaktadır. Bunun yanı sıra, arayüzü istatistik ve bilgisayar kullanımı konusunda yeterli bilgi ve deneyimi olmayan kullanıcılara uygun bir şekilde yapılandırılmıştır. Bir "benzetim-destekli yazılım" olan GAYUS'un "Yapay Us yardımlı benzetim" gurubunda yer alan "bilen ön arayüz" türüne örnek olduğu söylenebilir. Ayrıca, geliştirilen sistemin başarımı Monte Carlo yöntemi ile türetilen rasgele örneklemler üzerinde bir grup kullanıcı yardımı ile sınanmıştır. Bu kullanıcılar tarafından anlaşılır ve kolay kullanılır olarak değerlendirilen GAYUS'un, örneklem ölçümü büyüdükçe verinin türetildiği teorik dağılımı en iyi model olarak önerdiği gözlenmiştir. Sistem başarımının daha çok sayıda kıstas ve daha güçlü girdi analizi yöntemleri kullanılarak artırılabileceği düşünülmektedir.

Kaynakça

  • BANKS, J. ( 1998a), Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice, New York: John Wiley and Sons.
  • BANKS, J. (1998b), Software for Simulation, J. Banks (Der), Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice İçinde, New York: John Wiley and Sons, 813-835.
  • CHAMBERS, J., CLEVELAND, W., KLEINER, B. ve TUKEY, P. (1983), Graphical Methods for Data Analysis, California: Wadsworth Int. Group.
  • D'AGOSTINO, R.B. ve STEPHENS. M.A. (1986), Goodness-of-Fit Techniques, New York: Marcel Dekker.
  • DANIEL, W. (1978), Applied Nonparametric Statistics, Boston: Houghton Mifflin Com.
  • ELIAS, M. (1996), Building Expert Systems: Principle Procedures and Applications, Minneapolis: Wadsworth .
  • FREUND, J.E. ( 1992), Mathematical Statistics. New Jersey: Prentice Hall.
  • GIARRATANO. J. ve RILEY. G. (1994). Expert Systems: Principles and Programming, Boston: PWS Pub.
  • GORDON, G. (1978), System Simulation New Jersey: Prentice Hall Inc .
  • GRAY, C. ve STRAIN, T. (1995), Visual Basic 4 Nuts and Bolts: for Experienced Programmers, California: McGraw Hill.
  • HAYES-ROTH, F., WATERMAN. A.D. ve LENAT, B.D. (1983), Building Expert Systems, New-York: Addison-Wesley.
  • HOAGLIN, D.C, MOSTELLER, F. ve TUKEY, J.W. (1983), Understanding Robust and Exploratory Data Analysis, Canada: John Wiley and Sons.
  • JOHNSON, N.L., KOTZ. S. ve KEMP, A.W. (1994), Continuous Univariate Distributions. Vol 1. Boston: Houghton Mifflin.
  • JOHNSON N.L., KOTZ. S. ve KEMP, AW. (1995). Continuous Univariate Distributions, Vol 2, Boston: Houghton Mifflin.
  • JOHNSON. N.L., KOTZ, S. ve KEMP. A.W. (1992), Univariate Discrete Distributions. Boston: Houghton Mifflin.
  • KELTON, W.D., SADOWSKl, R.P. ve SADOWSKI, D.A. (2002), Simulation with ARENA, Boston: McGraw-Hill.
  • KONVALINEN, J. ve WILEMAN. S. (1987). Programming with Pascal. Singapore: McGraw-Hill.
  • LAW. A.M. ve KELTON, W.D. (2000), Simulation Modeling and Analysis, New Jersey: McGraw-HiIl.
  • LAW. AM. ve McCOMAS, M.G. (1 999), ExpertFit: Total Support for Simulation Input Modeling, AP. Farrington. H.B. Newbhard. D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde, 261-266.
  • LEEMIS. L. (1999), Simulation Input Modelling, A.P. Farrington. H.B. Newbhard, D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde. 14-23.
  • MARIA, A. (1997), Introduction to Modeling and simulation, S. Andradottir, K.J. Healy, D.H. Withers ve B.L. Nelson (der.). Proceedings of Winter Simulation Conference içinde, 7-13.
  • NELSON, B.L. ve YAMNITSKY. M. (1998). Input Modelling Tools for Complex Problems. A.P. Farrington. H.B. Newbhard. D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulation Conference içinde. 105-111.
  • NILSEN, N.R. (1991). Application of AI Techniques to Simulation, P.A. Fishwick, P.A ve R.B. Modjeski (der.), Knowledge-Based Simulation Methodology and Application içinde, New York: Springer-Verlag, 1-19.
  • ÖREN, T. (1994), Artificial lntelligence in Simulation, Annals of Operations Research. 63, 287- 319.
  • ÖZTÜRK, A. ve DUDEWICZ, E.J. (1992), a New Statistical Goodness-of-fit Test Based on Graphical Representation, Biometrica, 34, 403-427.
  • PROLOG DEVELOPMENT CENTER (1993), User's Book for Esta for Window, Denmark: PDC
  • RICH, E. (1983), Artificial Intelligence, New Jersey: MacGraw-Hill.
  • ROTHENBERG, J. (1990), Tutorial: Artificle Intelligence and Simulation, O. Balci, R.P. Sadowski ve R.E. Nance (der), Proceedings of Winter Simulation Conference içinde, 22-24.
  • RYAN. B.F., JOINER. B.L. ve RYAN, Jr. T.A. (1985), Minitab Handbook., Boston: PWS-Kent.
  • SCHMEISER, B. (1999) Advanced input Modeling for Simulation Experimentation, A.P. Farrington, H.B. Newbhard, D.T. Sturrock ve G.W. Evans (der.), Proceedings of Winter Simulatian Conference içinde, 110- 115.
  • SELIA, A. (1995), Introduction to Simulation, C. Alexopoulos. K. Kang, W.R. Lilegdon ve D. Goldsman (der.), Proccedings of Winter Simulation Conference içinde. 7-14.
  • SHANNON, R.E. (1998), lntroduction to Art and Science of Simulation, DJ. Medeiros, E.F. Watson, J.S. Carson, M.S. Manivannan (der.), Proceedings of Winter Simulation Conference içinde, 7-14.
  • SHAPIRO, S.S. ve WILK, M.B. (1965), An Analysis of Variance Test for Normality, Biometrika, 52, 591-611.
  • TANIMOTO, S.L. ( 1995), The Elements of AI Using Common Lisp, Oxford: CS Press.
  • VINCENT, S. (1998), Input Data Analysis, J. Banks (der.), Handbook of Simulation: Principles, Methodology, Advances, Application, and Practice içinde, New York: John Wiley and Sons, 55-91.
  • YILMAZ, A. ve SABUNCUOĞLU, İ. (2000), lnput Data Analysis Using Neural Networks, Simulation,74(3),128-137.
Toplam 36 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular İstatistiksel Teori
Bölüm Araştırma Makaleleri
Yazarlar

Halil Tanıl

İnci Batmaz

Yayımlanma Tarihi 15 Aralık 2003
Yayımlandığı Sayı Yıl 2003 Cilt: 2 Sayı: 3

Kaynak Göster

APA Tanıl, H., & Batmaz, İ. (2003). Benzetimde Girdi Analizi Yapan Otomatik Bir Sistem. İstatistik Araştırma Dergisi, 2(3), 31-47.