|
The main objective of the present paper is to demonstrate the power of statistical time series for analyzing supply chain inventory data. Our specific goal addresses two critical management needs that relate to the development of training tools for new corporate hires to our audit management team and providing quantitative methodologies for capturing or recognizing the underlying nature of supply chain inventory data.
This current effort will involve developing several robust autoregressive integrated moving average (ARIMA) time series models that describe the dynamical behavior of our inventory data. We will highlight the efficacy of the resulting models using generated forecast estimates. An array of classical statistical metrics will also be used to evaluate the sensitivity and stability of generated forecast estimates. In addition we will provide suggestions about how such models may be used to improve audit verification, minimize total inventory cost and identify potential product shortages or demand change points.
|
