Conclusions

Predictive computer models have proven to effective in reducing processing consts and improving product quality in the food industry >These models also reduce the number of costly and time-consuming production trials needed for the development of new products or processes .Modelsare also successfully used for on-line control systems needed to maintain a constant product quality .The development of models for the food industry is an ongoing process.The integration of physico-chemical models ,computational fluid dynamics and control systems will enable an optimal use of availabe knowledge ,experimental data and comoputing speed .As a result ,it will soon be possible to predict more accurately the optimal process setings for the desired powder properties and to control the quality on-line.This will make it possible to adjust drying equipment quickly to newa products and will reduce variations in product properties to a large extent.