Pattern recognition method has been applied to solving the problems of the optimization of chemical processes. The basic starting-point of optimization by pattern recognition is that the technological parameters are used as feature variable to construct the pattern space and all samples are divided into a number of classes according to the production target to be optimized. In order to find out the key factors influencing the target

the method of feature extraction is adopted to reduce the dimensionality of the pattern space of technological parameters. Some specific methods are suggested to search for optimal production conditions for two kinds of optimization problems

i. e.

the problem of optimal index and the problem of optimal direction

respectively. The problems of multiple targets are also discussed briefly. The methods of process optimization by pattern recognition have been applied to some factories and considerable technical-economic benefits have been achieved.