Copyright © Cloud Software Group, Inc. All Rights Reserved
Copyright © Cloud Software Group, Inc. All Rights Reserved


Chapter 7 Cases : predict

predict
The predict table stores the prediction data for all expected work items currently defined on the system.
Structure
The predict table has the following structure:
TABLE predict (
node_id number(5) NOT NULL,
proc_num number(5) NOT NULL,
case_num number(20) NOT NULL,
parent_proc_num number(5) NOT NULL,
parent_case_num number(20) NOT NULL,
main_proc_num number(5) NOT NULL,
main_case_num number(20) NOT NULL,
step_name varchar2(8) NOT NULL,
step_desc varchar2(24) NULL,
step_desc2 varchar2(24) NULL,
step_addr varchar2(49) NOT NULL,
step_durn_secs number(10) NULL,
step_durn_usecs number(10) NULL,
step_start date NOT NULL,
step_start_usecs number(10) NOT NULL,
step_end date NOT NULL,
step_end_usecs number(10) NOT NULL,
field_name varchar2(31) NULL,
field_value varchar2(255) NULL)
0, if this is a predicted work item in a future sub-case, rather than in a currently outstanding sub-case.
0, if this is a predicted work item in a future sub-case, rather than a currently outstanding sub-case, that was itself started from a predicted future sub-case.
Note: The step_durn_usecs column can be combined with this column to provide resolution to a microsecond.
Note: The step_start_usecs column can be combined with this column to provide resolution to a microsecond.
Number of microseconds since the start of the seconds value specified in the step_start column.
Note: The step_end_usecs column can be combined with this column to provide resolution to a microsecond.
Number of microseconds since the start of the seconds value specified in the step_end column.
Value of the CDQP assigned to the field_name field for this predicted work item.
Primary Key
None.
Foreign Keys
The following foreign key is defined for this table.

1
This key enforces the DELETE CASCADE referential action.

Indexes
None.
Storage
The following STORAGE values are defined for this table.
By default, the predict and predict_lock tables use the SMALL macro values. If you intend to enable case prediction on your system, TIBCO recommends that you change these tables to use a larger value in line with the level of prediction activity you expect.
Table Activity
The predict table contains one or more rows for each predicted work item generated by each step of each case of each procedure that currently has prediction data defined for it.
If a predicted work item contains one or more fields that have CDQPs assigned to them, duplicate rows are added for each CDQP. In the first row, the field_name and field_value columns are blank. Each subsequent row contains the field_name and field_value entries for one assigned CDQP. For example, if a predicted work item contains 5 fields that have CDQPs assigned to them, it will have 6 rows in this table.
Rows are added, updated and deleted in the following situations.
Note: One row is added for each step in the procedure that can occur on the currently predicted path(s).
Note: All rows for a given main case number are deleted for each step in the procedure that can no longer occur on the currently predicted path(s).
Case prediction can be enabled and disabled using the ENABLE_CASE_PREDICTION process attribute. See TIBCO iProcess Engine Administrator's Guide for more information.

Copyright © Cloud Software Group, Inc. All Rights Reserved
Copyright © Cloud Software Group, Inc. All Rights Reserved