Forecasting Air-to-Ground Munitions Demand in Military Operations Using Markov-Modulated Poisson Process
In this paper, we examine the problem of modeling and forecasting the air-to-ground munitions demand in military operations. We consider a continuous state space time series approach and we model the munitions demand as a non-homogeneous Poisson process with intensity function that depends on a continuous-time Markovian operational tempo process. This approach is known in stochastic process as a Markov-Modulated Poisson Process (MMPP) and has successfully been applied to queuing and communication network problems. We consider four operational tempo states, corresponding to the different intensity levels of military operations. We apply the maximum likelihood estimation method to derive the MMPP model parameters, assuming that the Markov chain states are observed. We illustrate the method with an example using two time series data sets of different trends. This study provides military planners with a decision support method for forecasting munitions requirements in the face of abrupt changes in demand. An extension of the model would be to use Bayesian approach for inferring the MMPP parameters through application of a Forward-Backward type algorithm to capture potential hidden states.
1. Andrews, W.S; Hurley, W.J. Approaches to Determining Army Operational Stockpile Levels. Canadian Military Journal, 2004, 5(2), 34-46.
2. Bean, W.L. Inventory Management under Uncertainty: a Military Application; University of Pretoria, 2011.
3. Box, G.E.P.; Jenkins, G.M.; Reinsel, G.C. Time Series Analysis: Forecasting and Control; John Wiley & Sons Inc., New Jersey, 2008.
4. Nenni, M.E.; Giustiniano, L.; Pirolo, L. Demand Forecasting in the Fashion Industry: a Review. International Journal of Engineering Business Management, Special Issue on Innovations in Fashion Industry, 2013, 5(37), 1-6.
5. Durban, J.; Koopman, S.J. Time Series Analysis by State Space Methods; Oxford University Press, 2012.
6. Koller, D.; Friedman, N. Probabilistic Graphical Models; Cambridge, MA: MIT Press, 2009.
7. Cox, D.R. Some Statistical Methods Connected with Series of Events. Journal of the Royal Statistical Society: Series B, 1955, 17, 129–164.
8. Fischer, W.; Meier-Hellstern, K. The Markov-Modulated Poisson Process (MMPP) Cookbook. Performance Evaluation, 1993, 18 (2), 149-171.
9. Muscariello, L.; Mellia, M.; Meo, M.; Ajmone, M.; Cigno, R.L. An MMPP-Based Hierarchical Model of Internet Traffic. IEEE Communications Society, 2004, 2143-2147.
10. Lu, S. Markov Modulated Poisson Process Associated with State-Dependent Marks and its Applications to the Deep Earthquakes. Annals of the Institute of Statistical Mathematics, 2012, 64, 87-106.
11. Lin, C.; Kim, M.J.; Makis, V. A Stochastic Model for Highway Accident Predictions with Winter Data. The Open Statistics and Probability Journal, 2013, 5, 6-13.
12. Thayakaran, R.; Ramesh, N.I. Markov Modulated Poisson Process Models Incorporating Covariates for Rainfall Intensity. Water Science and Technology: a Journal of the International Association on Water Pollution Research, 2013, 67(8), 1786-1792.
13. Royal Air Force, Aircraft and Weapons; Jay Myers, Royal Air Force Media and Communications, Headquarters Air Command, 2013.
14. Rydén, T. An EM Algorithm for Estimation in Markov-Modulated Poisson Processes. Computational Statistics & Data Analysis, 1996, 21, 431-447.
15. Aggarwal, C.C.; Reddy, C.K. Data Clustering: Algorithms and Applications; Chapman and Hall/CRC, 2013.
16. Fearnhead, P.; Sherlock, C. An Exact Gibbs Sampler for the Markov-Modulated Poisson Process. Journal of the Royal Statistical Society: Series B, 2006, 68 (5), 767-784.
17. Geyer, C.Y. Practical Markov Chain Monte Carlo. Statistical Science, 1992, 7, 473-483.
18. Bowerman, B.L.; O'Connell, R.T.; Koehler, A.B. Forecasting, Time Series and Regression, 4th edition; Belmont CA: Thomson Brooks/Cole, 2005.