Article ID: | iaor2001171 |
Country: | Netherlands |
Volume: | 120 |
Issue: | 1 |
Start Page Number: | 108 |
End Page Number: | 117 |
Publication Date: | Jan 2000 |
Journal: | European Journal of Operational Research |
Authors: | Natter Martin, Feurstein Markus |
Keywords: | neural networks, programming: dynamic |
The valuation of Flexible Manufacturing Systems is one of the most frequently undertaken productivity improvement activities. In practice, the introduction of an FMS into industry must be done on the basis of cost justification. Recently developed techniques for the evaluation of the value of flexibility typically include the computation of stochastic dynamic programs. However, the computational effort of stochastic dynamic programs grows combinatorially and limits application to real world problems. In this contribution we derive fast approximations to the stochastic dynamic program and compare their results to the exact solution. The proposed methods show an excellent worst case behavior (1%) for a wide range of volatility of the underlying stochastic profit margins and costs for switching the production mode. The computational effort is reduced by a factor of more than 200.