Probabilistic Methods for Long-Term Demand Forecasting for Aviation Production Planning

Supplemental Files
Date
2017-01-01
Authors
Jackman, John
MacKenzie, Cameron
Zhang, Minxiang
MacKenzie, Cameron
Hu, Guiping
Krejci, Caroline
Jackman, John
Hu, Guiping
Hu, Charles
Burnett, Gabriel
Graunke, Adam
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Journal Issue
Series
Department
Industrial and Manufacturing Systems EngineeringBioeconomy Institute (BEI)
Abstract

The aviation industry represents a complex system with low-volume high-value manufacturing, long lead times, large capital investments, and highly variable demand. Making important decisions with intensive capital investments requires accurate forecasting of future demand. However, this can be challenging because of significant variability in future scenarios. The use of probabilistic methods such as Brownian motion in forecasting has been well studied especially in the financial industry. Applying these probabilistic methods to forecast demand in the aerospace industry can be problematic because of the independence assumptions and no consideration of production system in these models. We used two forecasting models based on stochastic processes to forecast demand for commercial aircraft models. A modified Brownian motion model was developed to account for dependency between observations. Geometric Brownian motion at different starting points was used to accurately account for increasing variation. The modified Brownian motion and the geometric Brownian motion models were used to forecast demand for aircraft production in the next 20 years.

Comments

This proceeding is published as Zhang, Minxiang, Cameron A. MacKenzie, Caroline Krejci, John Jackman, Guiping Hu, Charles Y. Hu, Gabriel A. Burnett, and Adam A. Graunke. "Probabilistic methods for long-term demand forecasting for aviation production planning." In Proceedings of the 2017 IISE Annual Conference and Expo. May 20-23, 2017. Pittsburgh, Pennsylvania. Posted with permission.

Description
Keywords
Citation
DOI
Source