Inspection result analysis device, inspection result analysis method, and storage medium
An inspection result analysis device includes a processor. The processor factorizes each first distribution of a first distribution group into a second distribution group and a weight group corresponding to the second distribution group by non-negative matrix factorization of the first distribution group. Each first distribution is a distribution of numerical data on a surface of a substrate, the numerical data resulting from an inspection of the substrate. The processor generates a fourth distribution by multiplying a third distribution by a weight corresponding to the third distribution to acquire a representative value of a numerical data group constituting the fourth distribution. The processor associates a fifth distribution with the third distribution when the representative value exceeds a threshold. The fifth distribution is one first distribution of the first distribution group. The third distribution is one second distribution of the second distribution group which is factorized from the fifth distribution.
1. An inspection result analysis device comprising:
a memory; and
a processor connected to the memory, the processor configured to:
factorize each first distribution of a first distribution group into a second distribution group and a weight group by non-negative matrix factorization of the first distribution group, each first distribution of the first distribution group being a distribution of numerical data on a surface of a substrate, the numerical data resulting from an inspection of the substrate, each weight of the weight group corresponding to a respective second distribution of the second distribution group;
generate a fourth distribution by multiplying a third distribution by a weight corresponding to the third distribution;
acquire a representative value of numerical data group constituting the fourth distribution; and
associate a fifth distribution with the third distribution when the representative value exceeds a threshold, and refrain from associating the fifth distribution with the third distribution when the representative value is less than the threshold, the fifth distribution being one first distribution of the first distribution group, the third distribution being one second distribution of the second distribution group which is factorized from the fifth distribution.
2. The inspection result analysis device according to claim 1 , wherein the processor is further configured to:
sequentially select each first distribution of the first distribution group as the fifth distribution,
sequentially select each second distribution of the second distribution group as the third distribution upon selecting the fifth distribution, and
determine an association between each pair of the selected fifth distribution and the selected third distribution with reference to the threshold.
3. The inspection result analysis device according to claim 2 , wherein
the processor is further configured to output a pair of a sixth distribution and one or more first distributions associated with the sixth distribution, the sixth distribution being one second distribution of the second distribution group.
4. The inspection result analysis device according to claim 3 , wherein
the processor is further configured to count a number of the one or more first distributions associated with the sixth distribution, and output a result of the counting in association with the pair of the sixth distribution and the one or more first distributions.
5. The inspection result analysis device according to claim 1 , wherein
the representative value corresponds a maximum value, an average value, a median value, an upper limit value of a 95% confidence interval, or a third quartile of the numerical data group constituting the fourth distribution, or a value obtained by multiplying any of the maximum value, the average value, the median value, the upper limit value of the 95% confidence interval, and the third quartile by a constant.
6. The inspection result analysis device according to claim 1 , wherein
the numerical data constituting each first distribution is a value obtained in units of chips formed on the substrate.
7. The inspection result analysis device according to claim 6 , wherein
the numerical data constituting each first distribution is a number of defects obtained by the inspection of each of the chips.
8. A non-transitory, tangible computer-readable storage medium storing programmed instructions that, when executed by a computer, cause the computer to execute:
factorizing each first distribution of a first distribution group into a second distribution group and a weight group by non-negative matrix factorization of the first distribution group, each first distribution of the first distribution group being a distribution of numerical data on a surface of a substrate, the numerical data resulting from an inspection of the substrate, each weight of the weight group corresponding to a respective second distribution of the second distribution group;
generating a fourth distribution by multiplying a third distribution by a weight corresponding to the third distribution;
acquiring a representative value of numerical data group constituting the fourth distribution; and
associating a fifth distribution with the third distribution when the representative value exceeds a threshold while refraining from associating the fifth distribution with the third distribution when the representative value is less than the threshold, the fifth distribution being one first distribution of the first distribution group, the third distribution being one second distribution of the second distribution group which is factorized from the fifth distribution.
9. The storage medium according to claim 8 , wherein
the programmed instructions cause the computer to further execute:
sequentially selecting each first distribution of the first distribution group as the fifth distribution;
sequentially selecting each second distribution of the second distribution group as the third distribution upon selecting the fifth distribution; and
determining an association between each pair of the selected fifth distribution and the selected third distribution with reference to the threshold.
10. The storage medium according to claim 9 , wherein
the programmed instructions cause the computer to further execute:
outputting a pair of a sixth distribution and one or more first distributions associated with the sixth distribution, the sixth distribution being one second distribution of the second distribution group.
11. The storage medium according to claim 10 , wherein
the programmed instructions cause the computer to further execute:
counting a number of the one or more first distributions associated with the sixth distribution; and
outputting a result of the counting in association with the pair of the sixth distribution and the one or more first distributions.
12. The storage medium according to claim 8 , wherein
the representative value corresponds a maximum value, an average value, a median value, an upper limit value of a 95% confidence interval, or a third quartile of the numerical data group constituting the fourth distribution, or a value obtained by multiplying any of the maximum value, the average value, the median value, the upper limit value of the 95% confidence interval, and the third quartile by a constant.
13. The storage medium according to claim 8 , wherein
the numerical data constituting each first distribution is a value obtained in units of chips formed on the substrate.