Method for decision making in a semiconductor manufacturing process

A method for constructing a decision model for making a decision within a semiconductor manufacturing process, the method comprising:

• obtaining data relating to one or more parameters of a patterning step of the semiconductor manufacturing process;
• deriving one or more categorical indicators from the data, each of the one or more categorical indicators being indicative of a quality of the semiconductor manufacturing process based on at least one threshold value;
• and
• training, by a hardware computer system, the decision model on measurements performed on substrates having undergone the patterning step and corresponding decisions and/or categories assigned thereto, the training setting a value for each of the one or more threshold values, so that the decision model outputs an appropriate value for each of the one or more categorical indicators based on the data.

A computer program product comprising a non-transitory computer-readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:

• obtain values of functional indicators derived from data generated during one or more operations of a semiconductor manufacturing process on a substrate, the functional indicators characterizing at least one operation;
• apply a decision model comprising one or more threshold values to the values of the functional indicators to obtain one or more categorical indicators;
• assign a category to the substrate based on the one or more categorical indicators;
• and
• use the category assigned to the substrate to assign a decision to the substrate, the decision being one of:
• i) rework the substrate, ii) inspect the substrate, or iii) use the substrate in a subsequent process step,
• wherein the decision model is trained on measurements performed on processed substrates having corresponding decisions and/or categories assigned thereto, the training setting at least one of the one or more threshold values.

A computer program product comprising a non-transitory computer-readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:

• obtain data relating to one or more parameters of a patterning step of a semiconductor manufacturing process;
• derive one or more categorical indicators from the data, each of the one or more categorical indicators being indicative of a quality of the semiconductor manufacturing process based on at least one threshold value;
• and
• train a decision model for making a decision within the semiconductor manufacturing process, on measurements performed on substrates having undergone the patterning step and corresponding decisions and/or categories assigned thereto, the training setting a value for each of the one or more threshold values, so that the decision model outputs an appropriate value for each of the one or more categorical indicators based on the data.