Reduced simulation model for flow analysis in a sawmill internal supply chain

Accurate and simple simulation flow model is useful to make decision at the right time to manage supply chain or workshop. To do that, different reduction model complexity approaches have been proposed. One of them is to associate discrete event model of bottlenecks and continuous model of other work centers according to the theory of constraints. The continuous approximation is used only do determine how the bottlenecks are fed. Different continuous model have been proposed in the past. This paper focuses on the association of regression trees and neural networks in order to benefit of the advantages of each other. This approach is used for the modeling of a sawmill workshop and the results are compared with those obtained previously by using only CART model or neural network model.

[1]  Stephen Ward,et al.  Arguments for Constructively Simple Models , 1989 .

[2]  Bernard Widrow,et al.  Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[3]  André Thomas,et al.  Accepté à : Engineering Applications of Artificial Intelligence 1 Multilayer Perceptron for Simulation Models Reduction : Application to a Sawmill Workshop , 2017 .

[4]  Marie-Christine Suhner,et al.  A New Multilayer Perceptron Pruning Algorithm for Classification and Regression Applications , 2014, Neural Processing Letters.

[5]  William L. Berry,et al.  Manufacturing Planning and Control Systems , 1984 .

[6]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[7]  K. Narasimhan,et al.  The Goal: A Process of Ongoing Improvement , 2005 .

[8]  F. Robert Jacobs,et al.  MANUFACTURING PLANNING AND CONTROL SYSTEMS FOR SUPPLY CHAIN MANAGEMENT , 2004 .

[9]  Gérard Bloch,et al.  Neural Modeling of an Induction Furnace Using Robust Learning Criteria , 1999, Integr. Comput. Aided Eng..

[10]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[11]  André Thomas,et al.  A neural network for the reduction of a product-driven system emulation model , 2011 .

[12]  Roger J. Brooks,et al.  Simplification in the simulation of manufacturing systems , 2000 .

[13]  Jun-Fei Qiao,et al.  A structure optimisation algorithm for feedforward neural network construction , 2013, Neurocomputing.

[14]  Moutaz Khouja,et al.  An aggregate production planning framework for the evaluation of volume flexibility , 1998 .

[15]  André Thomas,et al.  CART for Supply Chain Simulation Models Reduction - Application to a Sawmill Internal Supply Chain , 2014, APMS.

[16]  Ray J. Paul,et al.  Discrete event simulation model reduction: A causal approach , 2006, Simul. Model. Pract. Theory.

[17]  Patrick Charpentier,et al.  Reducing simulation models for scheduling manufacturing facilities , 2005, Eur. J. Oper. Res..

[18]  Bernard P. Zeigler,et al.  Theory of Modelling and Simulation , 1979, IEEE Transactions on Systems, Man and Cybernetics.

[19]  Russell M. Tahmoush,et al.  Simulation for planning and scheduling , 1970 .

[20]  R. Lewis.  An Introduction to Classification and Regression Tree (CART) Analysis , 2000 .