Validation and Selection between Machine Learning Technique and Traditional Methods to Reduce Bullwhip Effects: a Data Mining Approach
The aim of this paper is to present a methodology in
three steps to forecast supply chain demand. In first step, various data
mining techniques are applied in order to prepare data for entering
into forecasting models. In second step, the modeling step, an
artificial neural network and support vector machine is presented
after defining Mean Absolute Percentage Error index for measuring
error. The structure of artificial neural network is selected based on
previous researchers' results and in this article the accuracy of
network is increased by using sensitivity analysis. The best forecast
for classical forecasting methods (Moving Average, Exponential
Smoothing, and Exponential Smoothing with Trend) is resulted based
on prepared data and this forecast is compared with result of support
vector machine and proposed artificial neural network. The results
show that artificial neural network can forecast more precisely in
comparison with other methods. Finally, forecasting methods'
stability is analyzed by using raw data and even the effectiveness of
clustering analysis is measured.
[1] C. W. J. Granger, "Can we improve the perceived quality of economic
forecasts?", J. Appl. Econom.,. Vol. 11, 1996, pp. 455-473.
[2] P. H. Franses, G. Draisma, "Recognizing changing seasonal patterns
using artificial neural networks", J. Econometrics, Vol. 81, 1997, pp.
273-280.
[3] R. T. Peterson, "Forecasting practices in the retail industry", J. Bus.
Forec., vol. 12, 1993, pp. 11-14.
[4] M. Qi, G. S. Maddala, "Economic factors and the stock market: a new
perspective", J. Forec., vol. 18 no. 3, 1999, pp. 151-166.
[5] M. Qi, "Nonlinear predictability of stock returns using financial and
economic variables", J. Bu. Econom. Statist., vol. 17, no. 4, 1999, pp.
419-429.
[6] M. Qi, "Financial applications of artificial neural networks", in
Handbook of Statistics, Statistical Methods in Finance, vol. 4, G. S.
Maddala, C. R. Rao, Ed. North-Holland, Elsevier Science Publishers,
Amsterdam, 1996, pp. 529-552.
[7] K. A. Krycha, U. Wagner, "Applications of artificial neural networks in
management science: a survey", J. Retail. Cons. Ser., vol. 6, 1999, pp.
185-203.
[8] G. P. Zhang, B. E. Patuwo, M. Y. Hu, "Forecasting with artificial neural
networks: the state of the art", Intl. J. Forec., vol. 14, no. 1, 1998, pp.
35-62.
[9] B. Jeong, H. Jung, N. Park, "A computerized causal forecasting system
using genetic algorithms in supply chain management", J. Sys. Software,
vol. 60, 2002, pp. 223-237.
[10] T. R. Willemain, C. N. Smart, H. F. Schwarz, "A new approach to
forecasting intermittent demand for service parts inventories", Intl. J.
Forec., vol. 20, 2004, pp. 375- 387.
[11] C. S. Hilas, S. K. Goudos, J. N. Sahalos, "Seasonal decomposition and
forecasting of telecommunication data: A comparative case study",
Technol. Forec. Soc. Change, vol. 73, 2006, pp. 495-509.
[12] S. Kang, "An investigation of the use of feedforward neural networks for
forecasting", Ph.D. Dissertation, Kent State University, Kent, Ohio,
1991.
[13] T. Hill, M. O'Connor, W. Remus, "Neural network models for time
series forecast", Manag. Sci., vol. 42, no. 7, 1996, pp. 1082-1092.
[14] M. Nelson, T. Hill, B. Remus, M. O'Connor, "Can neural networks be
applied to time series forecasting learn seasonal patterns: an empirical
investigation", in Proc. 27th Ann. Int. Conf. System Sciences, Hawaii,
1994, pp. 649-655.
[15] B. Foster, F. Collopy, L. Ungar, "Neural network forecasting of short,
noisy time series", Comput. Chem. Eng., vol. 16, no. 12, 1992, pp. 293-
297.
[16] M. Dugan, K. A. Shriver, A. Peter, "How to forecast income statement
items for auditing purposes", J. Bus. Forec., vol. 13, 1994, pp. 22-26.
[17] I. Alon, "Forecasting aggregate retail sales: the Winters' model
revisited", in The 1997 Annual Proceedings, J. C. Goodale, Ed.,
Midwest Decision Science Institute, 1997, pp. 234-236.
[18] I. Alon, Q. Min, R. Sadowski, "Forecasting aggregate retail sales: a
comparison of artificial neural networks and traditional method", J.
Retail. Cons. Ser., vol. 8, 2001, pp. 147-156.
[19] T. M. O'Donovan, Short Term Forecasting: An Introduction to the Box-
Jenkins Approach. NY: New York, Wiley, 1983.
[20] R. Sharda, R. Patil, "Connectionist approach to time series prediction: an
empirical test", J. Intell. Manufac., vol. 3, 1992, pp.317-323.
[21] Z. Tang, C. de Almeida, P. Fishwick, "Time series forecasting using
neural networks vs. Box-Jenkins methodology", Simulation, vol. 57, no.
5, 1990, pp.303-310.
[22] V. Vapnik, The Nature of Statistical Learning Theory. New York,
Springer, 1995.
[23] V. Vapnik, S. E. Golowich, A. Smola, "Support vector method for
function approximation, regression estimation, and signal processing",
Adv. Neural inf. Process. Syst., vol. 9, 1997, pp. 287-291.
[24] S. Mukherjee, E. Osuna, F. Girosi, "Nonlinear prediction of chaotic time
series using support vector machines", in Proc. IEEE NNSP, 1997.
[25] S. Ruping, K. Morik, "Support Vector Machines and Learning about
Time", in Proc. of ICASSP, 2003.
[26] J. Han, M. Kamber, Data Preprocessing, Data Mining, concepts and
techniques, 2nd ed., Morgan Kaufman Publisher, 2006, ch2.
[27] T. Calinski, J. Harabasz, "A Dendrite Method for Cluster Analysis",
Commun. Statist., Vol. 3, 1974, pp. 1-27.
[28] K. Hornik, M. Stinchcombe, H. White, "Multilayer feedforward
networks are universal approximators", Neural Net., vol. 2, 1989, pp.
359-366.
[29] K. Hornik, M. Stinchcombe, H. White, "Universal approximation of an
unknown mapping and its derivatives using multilayer feedforward
networks", Neural Net., vol. 3, 1990, pp. 551-560.
[30] H. White, "Connectionist nonparametric regression: multilayer
feedforward networks can learn arbitrary mappings", Neural Net., vol. 3,
1990, pp. 535-549.
[31] H. Demuth, M. Beale, Neural Network Toolbox User's Guide, Version
3.0. The Math Works, Inc., 1997, pp. 5-35.
[32] D. J. C. MacKay, "Bayesian interpolation", Neural Comput., vol. 4,
1992, pp. 415-447.
[33] V. N. Vapnik, Statistical Learning Theory. NY: New York, John Wiley
& Sons, 1998.
[34] G. L. Lilien, P. Kotler, Marketing Decision Making: A Model Building
Approach. NY: New York, Harper and Row Publishers, 1983.
[1] C. W. J. Granger, "Can we improve the perceived quality of economic
forecasts?", J. Appl. Econom.,. Vol. 11, 1996, pp. 455-473.
[2] P. H. Franses, G. Draisma, "Recognizing changing seasonal patterns
using artificial neural networks", J. Econometrics, Vol. 81, 1997, pp.
273-280.
[3] R. T. Peterson, "Forecasting practices in the retail industry", J. Bus.
Forec., vol. 12, 1993, pp. 11-14.
[4] M. Qi, G. S. Maddala, "Economic factors and the stock market: a new
perspective", J. Forec., vol. 18 no. 3, 1999, pp. 151-166.
[5] M. Qi, "Nonlinear predictability of stock returns using financial and
economic variables", J. Bu. Econom. Statist., vol. 17, no. 4, 1999, pp.
419-429.
[6] M. Qi, "Financial applications of artificial neural networks", in
Handbook of Statistics, Statistical Methods in Finance, vol. 4, G. S.
Maddala, C. R. Rao, Ed. North-Holland, Elsevier Science Publishers,
Amsterdam, 1996, pp. 529-552.
[7] K. A. Krycha, U. Wagner, "Applications of artificial neural networks in
management science: a survey", J. Retail. Cons. Ser., vol. 6, 1999, pp.
185-203.
[8] G. P. Zhang, B. E. Patuwo, M. Y. Hu, "Forecasting with artificial neural
networks: the state of the art", Intl. J. Forec., vol. 14, no. 1, 1998, pp.
35-62.
[9] B. Jeong, H. Jung, N. Park, "A computerized causal forecasting system
using genetic algorithms in supply chain management", J. Sys. Software,
vol. 60, 2002, pp. 223-237.
[10] T. R. Willemain, C. N. Smart, H. F. Schwarz, "A new approach to
forecasting intermittent demand for service parts inventories", Intl. J.
Forec., vol. 20, 2004, pp. 375- 387.
[11] C. S. Hilas, S. K. Goudos, J. N. Sahalos, "Seasonal decomposition and
forecasting of telecommunication data: A comparative case study",
Technol. Forec. Soc. Change, vol. 73, 2006, pp. 495-509.
[12] S. Kang, "An investigation of the use of feedforward neural networks for
forecasting", Ph.D. Dissertation, Kent State University, Kent, Ohio,
1991.
[13] T. Hill, M. O'Connor, W. Remus, "Neural network models for time
series forecast", Manag. Sci., vol. 42, no. 7, 1996, pp. 1082-1092.
[14] M. Nelson, T. Hill, B. Remus, M. O'Connor, "Can neural networks be
applied to time series forecasting learn seasonal patterns: an empirical
investigation", in Proc. 27th Ann. Int. Conf. System Sciences, Hawaii,
1994, pp. 649-655.
[15] B. Foster, F. Collopy, L. Ungar, "Neural network forecasting of short,
noisy time series", Comput. Chem. Eng., vol. 16, no. 12, 1992, pp. 293-
297.
[16] M. Dugan, K. A. Shriver, A. Peter, "How to forecast income statement
items for auditing purposes", J. Bus. Forec., vol. 13, 1994, pp. 22-26.
[17] I. Alon, "Forecasting aggregate retail sales: the Winters' model
revisited", in The 1997 Annual Proceedings, J. C. Goodale, Ed.,
Midwest Decision Science Institute, 1997, pp. 234-236.
[18] I. Alon, Q. Min, R. Sadowski, "Forecasting aggregate retail sales: a
comparison of artificial neural networks and traditional method", J.
Retail. Cons. Ser., vol. 8, 2001, pp. 147-156.
[19] T. M. O'Donovan, Short Term Forecasting: An Introduction to the Box-
Jenkins Approach. NY: New York, Wiley, 1983.
[20] R. Sharda, R. Patil, "Connectionist approach to time series prediction: an
empirical test", J. Intell. Manufac., vol. 3, 1992, pp.317-323.
[21] Z. Tang, C. de Almeida, P. Fishwick, "Time series forecasting using
neural networks vs. Box-Jenkins methodology", Simulation, vol. 57, no.
5, 1990, pp.303-310.
[22] V. Vapnik, The Nature of Statistical Learning Theory. New York,
Springer, 1995.
[23] V. Vapnik, S. E. Golowich, A. Smola, "Support vector method for
function approximation, regression estimation, and signal processing",
Adv. Neural inf. Process. Syst., vol. 9, 1997, pp. 287-291.
[24] S. Mukherjee, E. Osuna, F. Girosi, "Nonlinear prediction of chaotic time
series using support vector machines", in Proc. IEEE NNSP, 1997.
[25] S. Ruping, K. Morik, "Support Vector Machines and Learning about
Time", in Proc. of ICASSP, 2003.
[26] J. Han, M. Kamber, Data Preprocessing, Data Mining, concepts and
techniques, 2nd ed., Morgan Kaufman Publisher, 2006, ch2.
[27] T. Calinski, J. Harabasz, "A Dendrite Method for Cluster Analysis",
Commun. Statist., Vol. 3, 1974, pp. 1-27.
[28] K. Hornik, M. Stinchcombe, H. White, "Multilayer feedforward
networks are universal approximators", Neural Net., vol. 2, 1989, pp.
359-366.
[29] K. Hornik, M. Stinchcombe, H. White, "Universal approximation of an
unknown mapping and its derivatives using multilayer feedforward
networks", Neural Net., vol. 3, 1990, pp. 551-560.
[30] H. White, "Connectionist nonparametric regression: multilayer
feedforward networks can learn arbitrary mappings", Neural Net., vol. 3,
1990, pp. 535-549.
[31] H. Demuth, M. Beale, Neural Network Toolbox User's Guide, Version
3.0. The Math Works, Inc., 1997, pp. 5-35.
[32] D. J. C. MacKay, "Bayesian interpolation", Neural Comput., vol. 4,
1992, pp. 415-447.
[33] V. N. Vapnik, Statistical Learning Theory. NY: New York, John Wiley
& Sons, 1998.
[34] G. L. Lilien, P. Kotler, Marketing Decision Making: A Model Building
Approach. NY: New York, Harper and Row Publishers, 1983.
@article{"International Journal of Mechanical, Industrial and Aerospace Sciences:54056", author = "Hamid R. S. Mojaveri and Seyed S. Mousavi and Mojtaba Heydar and Ahmad Aminian", title = "Validation and Selection between Machine Learning Technique and Traditional Methods to Reduce Bullwhip Effects: a Data Mining Approach", abstract = "The aim of this paper is to present a methodology in
three steps to forecast supply chain demand. In first step, various data
mining techniques are applied in order to prepare data for entering
into forecasting models. In second step, the modeling step, an
artificial neural network and support vector machine is presented
after defining Mean Absolute Percentage Error index for measuring
error. The structure of artificial neural network is selected based on
previous researchers' results and in this article the accuracy of
network is increased by using sensitivity analysis. The best forecast
for classical forecasting methods (Moving Average, Exponential
Smoothing, and Exponential Smoothing with Trend) is resulted based
on prepared data and this forecast is compared with result of support
vector machine and proposed artificial neural network. The results
show that artificial neural network can forecast more precisely in
comparison with other methods. Finally, forecasting methods'
stability is analyzed by using raw data and even the effectiveness of
clustering analysis is measured.", keywords = "Artificial Neural Networks (ANN), bullwhip effect,
demand forecasting, Support Vector Machine (SVM).", volume = "3", number = "1", pages = "33-7", }