Determining supply chain flexibility using statistics and neural networks : a comparative study

The purpose of this paper is to examine the application of neural networks as a flexibility and performance measure in supplier-manufacturer activities. The dimensions of information exchange, supplier integration, product delivery, logistics, and organisational structure are used as determinants factors affecting this supply chain flexibility. The data set was collected from more than 200 Australian manufacturing firms evaluating their suppliers. Our study shows that neural networks can accurately determine a supplier’s flexibility with an error within 1%, which is more accurate than the conventional multivariate regression can.