Return to search

A systems dynamics perspective of forecasting in supply chains

Purpose: To evaluate the impact of forecasting on supply chain via a system dynamics perspective. Method/approach: Techniques from Control Theory (such as block diagram, z-transforms, Fourier transforms, Jury’s Inners approach, and frequency response analysis) and Time Series Analysis are used to investigate the performance of supply chains analytically. Simulation is also used to verify the results. Findings: This thesis provides a new and complete proof to the knowledge that Naïve, simple exponential smoothing, and Holt’s forecasting when used in the Order-Up-To (OUT) policy always produce the bullwhip effect for any demand pattern and for all lead-times. In terms of the bullwhip performance when Damped Trend (DT) forecasts are used in the OUT policy, the bullwhip effect is always generated for traditional parameter suggestions. However, the bullwhip avoidance behaviour occurs for some unconventional parameter values. Using these unconventional parameter values, the DT / OUT system acts like a low-pass filter that can eliminate the bullwhip effect and maintain good inventory performance at the same time. The thesis also proves that the Proportional Order-Up-To (POUT) policy is able to reduce system nervousness at the manufacturer. Moreover, the proportional future guidance (PFG) mechanism proposed may reduce system nervousness and inventory costs at the manufacturer and reduce the bullwhip effect in the supply chain simultaneously. Implications: This thesis shows that the bullwhip and net stock variance reduction behaviours exist when unconventional parameter values are used in the DT forecasting procedure. It is the first evidence that it is possible to design a system with good financial performance but without directly looking into the performance of forecasting. The thesis is also the first to consider the MRP nervousness problem and the bullwhip effect at the same time. The PFG method proposed is easy to understand, and since it does not require sophisticated integrated IT systems, or demand / inventory information sharing, it should be easy to implement.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:633584
Date January 2014
CreatorsLi, Qinyun
PublisherCardiff University
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://orca.cf.ac.uk/69349/

Page generated in 0.0014 seconds