System and method of data analysis

1. A method of data analysis, comprising:

comparing a data point in a data stream and a previous data point in the data;

stream to identify a change direction of the data point based on a difference between the data point and the previous data point;

comparing the change direction with a previous change direction to identify whether a trend change occurred for the data point;

identifying a change point that comprises a discrete data value in the data stream based on the trend change;

comparing the change point with discrete data; and

identifying a process alarm event that corresponds to the change point in the data stream.

2. The method of claim 1 , comprising calculating a Shewhart control chart value using the data point and previous data point to identify whether the change point exists based on a magnitude change.

3. The method of claim 1 , comprising calculating a cumulative sum value using the data point and previous data point to identify whether the change point exists based on a magnitude of change.

4. The method of claim 1 , comprising calculating a cumulative sum value using the data point and previous data point to identify whether the change point exists based on a magnitude of change only if a first technique does not detect that the change point exists.

5. The method of claim 1 , comprising identifying a relationship, if any, between the change point and a plurality of event occurrences.

6. The method of claim 1 , comprising identifying a relationship, if any, between a slope relating to the change point and a discrete event occurrence.

7. The method of claim 1 , comprising identifying a relationship, if any, between a change point and another change point.

8. The method of claim 1 , comprising categorizing the change point with a plurality of change points based on parameters of the change point similar to parameters of the plurality of change points.

9. The method of claim 8 , comprising categorizing the change point with the plurality of change points based on a time delay parameter of the change point.

10. The method of claim 1 , comprising:

marking the data point with a trend change marker if the trend change occurred;

and

identifying the change point in the data stream based on the trend change marker.

11. The method of claim 1 , comprising comparing the change point with a discrete event occurrence.

12. A system for data analysis, comprising:

a processor configured to operate a series of modules, wherein the modules comprise:

a change direction module adapted to compare a data point in a data stream and a previous data point in the data stream to identify a change direction of the data point based on a difference between the data point and the previous data point;

a trend change module adapted to compare the change direction with a previous change direction to identify whether a trend change occurred for the data point;

a change point module adapted to identify a change point that comprises a discrete data value in the data stream based on the trend change; and

an event module adapted to compare the change point with a discrete event occurrence.

13. The system of claim 12 , comprising a Shewhart control chart module adapted to calculate a Shewhart control chart value using the data point and previous data point to identify whether the change point exists based on a magnitude change.

14. The system of claim 12 , comprising a cumulative sum module adapted to calculate a cumulative sum value using the data point and previous data point to identify whether the change point exists based on a magnitude of change.

15. The system of claim 12 , comprising a cumulative sum module adapted to calculate a cumulative sum value using the data point and previous data point to identify whether the change point exists based on a magnitude of change only if the first technique does not detect that the change point exists.

16. The system of claim 12 , comprising a comparison module adapted to compare the change point to a plurality of event occurrences.

17. The system of claim 12 , comprising a slope comparison module adapted to compare a slope of the change point with the discrete event occurrence.

18. The system of claim 12 , comprising a change point comparison module adapted to compare the change point to another change point.

19. The system of claim 12 , comprising a categorization module adapted to categorize the change point with a plurality of change points based on parameters of the change point.

20. The system of claim 19 , wherein the categorization module is adapted to categorize the change point with the plurality of change points based on a time delay parameter of the change point.

21. The system of claim 12 , comprising a marking module adapted to mark the data point with a trend change marker if the trend change occurred.

22. The system of claim 21 , comprising an identification module adapted to identify the change point in the data stream based on the trend change marker.

23. A system for data analysis, comprising:

processing means for comparing a data point in a data stream and a previous data point in the data stream to identify a change direction of the data point based on a difference between the data point and the previous data point;

processing means for comparing the change direction with a previous change direction to identify whether a trend change occurred for the data point;

processing means for identifying a change point in the data stream based on the trend change; and

processing means for comparing the change point with other data.

24. A method of data analysis, comprising:

comparing a data point in a data stream and a previous data point in the data stream to identify a change direction of the data point based on a difference between the data point and the previous data point;

calculating a Shewhart control chart value using the data point and previous data point to identify whether a trend change occurred for the data point based on a magnitude of change;

marking the data point with a trend change marker if the trend change occurred;

identifying a change point in the data stream based on the trend change marker;

comparing the change point with a discrete event occurrence; and

identifying a process alarm event that corresponds to the change point in the data stream.

25. A method of data analysis, comprising:

comparing a first data point in a first data stream in a manufacturing process and a second data point in the first data stream in the manufacturing process to identify a change direction of the first data point based on a difference between the first data point and the second data point;

comparing the change direction of the first data point with a previous change direction relating to the first data stream in the manufacturing process to identify whether a trend change occurred for the first data point;

identifying a change point in the first data stream in the manufacturing process based on the trend change;

comparing a third data point in a second data stream in the manufacturing process and a fourth data point in the second data stream to identify a change direction of the third data point based on a difference between the third data point and the fourth data point;

comparing the change direction of the third data point with a previous change direction relating to the second data stream in the manufacturing process to identify whether a trend change occurred for the third data point;

identifying a change point in the second data stream in the manufacturing process based on the trend change;

comparing the change point in the first data stream in the manufacturing process with the change point in the second data stream in the manufacturing process; and

identifying a correlation between the change point in the first data stream in the manufacturing process and the change point in the second data stream in the manufacturing process.

26. A method of data analysis, comprising:

converting numeric data streams representing physical occurrences into data representing discrete events;

comparing the data representing discrete events to actual discrete events; and

identifying a correlation between the actual discrete events and the numeric data stream, wherein the discrete events represent process alarm events.