| Article ID: | iaor20117483 |
| Volume: | 9 |
| Issue: | 3 |
| Start Page Number: | 268 |
| End Page Number: | 279 |
| Publication Date: | Jul 2011 |
| Journal: | International Journal of Logistics Systems and Management |
| Authors: | Cannella Salvatore, Ciancimino Elena |
| Keywords: | systems, inventory, manufacturing industries, simulation, statistics: regression, statistics: decision, simulation: applications |
In the last 50 years, Forrester's system dynamics techniques have been adopted to analyse problems and find solutions for global supply chains. An important topic in production‐inventory system modelling is the design of experiment. The aim of this paper is to present an application of a statistical technique of design of experiment, the Latin Square Design, to set a combination of input values for the initial‐value problem of non‐linear repeated coupling of first‐order differential equations modelling a production‐inventory system. This design permits to reduce the number of experiments while allowing statistical analysis for testing the significance of the studied parameters.