This paper represents a long-standing collaboration between Robert Fildes and Brian Kingsman who sadly died before it could be completed. This paper proposes a methodology for examining the effect of demand uncertainty and forecast error on lot sizing methods, unit costs and customer service levels in MRP type manufacturing systems. A number of cost structures were considered which depend on th...

We present a new method of reducing the error in predicted wind speed, thus enabling better management of wind energy facilities. A numerical weather prediction model, COSMO, was used to produce 48 h forecast data every day in 2008 at horizontal resolutions of 10 and 3 km. A new adaptive statistical method was applied to the model output to improve the forecast skill. The method applied correct...

The purpose of this research is to the problem of optimal sizing of energy storage required for compensation of wind farm generation variability. Using wind farm production data from the BPA, we assess the effect of forecast quality and economic dispatch timing on the size of storage and critical power rating required to nearly perfectly match the committed energy. We develop a Model-Predictive...

The sensitivity of numerical weather forecasts to small changes in initial conditions is estimated using ensemble samples of analysis and forecast errors. Ensemble sensitivity is defined here by linear regression of analysis errors onto a given forecast metric. We show that ensemble sensitivity is proportional to the projection of the analysis-error covariance onto the adjoint sensitivity field...

This paper develops tourism demand econometric models based on the monthly data of tourists to Taiwan and adopts Multivariate Adaptive Regression Splines (MARS), Artificial Neural Network (ANN) and Support Vector Regression (SVR), MARS, ANN and SVR to develop forecast models and compare the forecast results. The results showed that SVR model is the optimal model, with a mean error rate of 3.61%...

This work is directed toward approximating the evolution of forecast error covariances for data assimilation. We study the performance of di erent algorithms based on simpli cation of the standard Kalman lter (KF). These are suboptimal schemes (SOS's) when compared to the KF, which is optimal for linear problems with known statistics. The SOS's considered here are several versions of optimal in...

Probabilistic Mesoscale Forecast Error Prediction Using Short-Range Ensembles

Driven by a trend towards renewable forms of generation, in particular wind, the nature of power system operation is changing. With respect to wind power, the uncertainty of the wind becomes an issue which must be considered. Through the use of wind forecasting, this uncertainty may be managed. The error inherent in forecasting will impact system reliability and cost as will inaccuracies in ass...

Despite the obvious desirability of formal testing procedures, earlier efforts at assessing the forecast accuracy of an estimated model revolved around the calculation of summary forecast error statistics e.g., see Klein (1991); Clements and Hendry (1993); Wallis (1995); Newbold, Harvey, and Leybourne (1999); and Stock and Watson (1999). Resort to such an informal approach may be ascribed mainl...

We propose a hybrid forecast model based on discrete grey-fuzzy Markov and grey –neural network model and show that our hybrid model can improve much more the performance of forecast than traditional grey-Markov model and neural network models. Our simulation results are shown that our hybrid forecast method with the combinational weight based on optimal grey relation degree method is better th...