Making use of the statistical simulation which is due to single step feedback the system in order you watch and to analyze production process and

摘要

(57) Abstract Statistical simulation of semiconductor production process is executed actual process in parallel. The input parameter which is obtained from probability density function is given by the simulator, the simulator this time, simulates the actual production process which is displayed in the form of probability density function. Repeating each simulation step and, calculating and using random seed value making use of the monte karuro method which is method of the trial and error which decides the best solution of problem, it is repeated. The simulator forms output with form of probability distribution. There is supplied from the actual inline measurement data in the simulation run statistical simulation making use of single step feedback, or it can use the input parameter which is obtained. The actual inline measurement data it can agree the both output data of the intermediate output structural data and the WET data which are formed by the simulator, the measurement data regarding the circumstance which is available. Probability density structure of the simulator is adjusted after each simulation step, the data which simulation is done that tries matches the inline measurement data more strictly.