Defining Data Extractions and Loading Datasets

To define data extractions, use the Define Extraction Criteria component (QS_QSX_DEF). This topic discusses how to define data extractions and load datasets.

Page Name

Definition Name

Usage

Define Extraction Criteria — General Page

QS_QSX_DEF

Define general settings for the extraction.

Define Extract Criteria — Selection Page

QS_QSX_FLDS

Define how to filter data for the extraction and select data-entry sessions.

Perform Data Extraction Page

QS_QSX_INQ

Execute the Data Extraction PeopleCode process.

Using the Quality application client, you can generate graphics and perform exploratory analysis of quality data while offline. The data managed by the PeopleSoft application client is called a dataset and contains the related measurement, traceability, attribute, and configuration information required for performing quality analysis. This information is formatted specifically for use by the Quality application client and resides outside the quality database.

Datasets are created as a result of defining and executing data extractions. The information taken from the Quality database fills a dataset based on selection and filtering criteria that you provide. Datasets are composed of spreadsheet-like elements, such as rows, columns, subsets (pages), and cell values. Datasets contain these options for column and row types:

  • Common columns (traceability).

    Common columns apply to all subsets in the dataset and usually represent traceability information, such as lot numbers and date stamps. Common columns help minimize redundant information when creating a standalone dataset.

  • Data columns (reading, defects, violations).

    Data columns are specific to an individual subset. Data columns typically include measurement values, subgroup counts, and attribute information, such as defects, causes, and actions.

    Note: Datasets created from the database, such as performing data extractions, use data columns exclusively.

  • Subset.

    Subset columns are logical groupings of data, typically based on a process stream or characteristic. A subset encompasses the columns in the dataset much like a page does in a typical spreadsheet program. Subsets are configured similarly to measurement plan characteristics; you can assign a data type and control parameters to a subset. This enables the subset to be graphed and analyzed easily.

    Note: Datasets created from the database populate the subset control configuration parameters with the current settings for the characteristic or stream. A dataset can contain up to 100 subsets. The tabs on the Quality application client spreadsheet represent subset names.

  • Standard cell.

    Standard cells contain a single value.

  • Tall cell.

    Tall cells contain multiple values. Attributes data, such as test violations, causes, and actions, can occur in multiples for a given cell, and thus may be displayed as multiple rows within the cell. Click the F2 button twice to enter multiple lines into a tall cell.

Datasets appear as a spreadsheet in the Quality application client. You create datasets using one of these methods:

  • Define and run a data extraction.

  • Access the Quality application client and create a dataset from scratch.

Extractions are reusable queries that retrieve data from the Quality database and can be shared by multiple users. The extraction functions (define and execute) are separated to facilitate access by users with different job functions, such as managers and analysts.

Use the Data Extraction PeopleCode process to select a set of inspection data from the PeopleSoft database and analyze it using the Quality application client tools. You define the data to extract and download it to the Quality application client for exploratory and ad-hoc analysis. Use the Quality application client to manipulate, add, change, or delete data to perform what-if analyses of processes and view graphics of the data.

Before defining a dataset for use by the Quality application client, you must:

  • Determine the processes to extract.

  • Decide how to organize the dataset.

  • Identify the categories of data to extract from the database.

  • Determine the amount of data to include in the extraction.

You must also use the using the File Locations page to enter the location to store the datasets.

See Defining Additional Common Information.

Field or Control

Description

Extraction Name

Enter an extraction name. An extraction identifies data you select from the Quality database used to create a dataset. Extraction names must conform to Microsoft Windows NT file naming conventions, so avoid using special characters, such as dashes, periods, and apostrophes in names.

Description

Enter the description for the subset. The name must be unique within the dataset.

Graph Preference ID

Select the default graph preference to associate with an extraction.

Use the Define Extraction Criteria - General page (QS_QSX_DEF) to define general settings for the extraction.

Navigation:

Quality > Prepare Analysis > Define Extraction Criteria > General

Field or Control

Description

Subset By

Select how the system subsets data for the extraction. Options are:

All (Single Subset): All data is displayed as one subset.

Characteristic: The data is divided into subsets according to individual characteristics.

Stream: The data is divided into subsets according to process stream.

Display Options

Select if the system should include one or all other categories of information to display, in addition to session data information.

Select Sessions By

Select how sessions are selected by timeframe. Options are:

All Subgroups: Extract all the subgroups for each process.

Daily Window: Extract the sessions entered in the last n days based on the current date and time. Enter the number of days to subtract from the current date in the Number of Days field.

Specific Date Range: Extract the sessions entered between a certain date range. Enter the date from and from time, and enter the date to and to time.

Subgroup Window: Extract the most current n subgroups entered. You'll enter the number of subgroups in the Number of Subgroups field.

Lock Selections

Select this check box to protect the integrity of the expressions and the amount of data to be included in the extraction. If selections aren't locked here, you can modify them during execution.

Use the Define Extraction Criteria - Selection Criteria page (QS_QSX_FLDS) to define how to filter data for the extraction and select data-entry sessions.

Navigation:

Quality > Prepare Analysis > Define Extraction Criteria > Selection Criteria

Specify the sessions from which you want to select data. You can limit the extraction to sessions entered for specific values of hierarchy data and trace data from the business unit and quality function, or you can limit data to sessions entered within a certain time period.

Field or Control

Description

Hierarchy Field Selections and Trace Field Selections

The fields vary depending on the quality function. Use these fields to further limit the sessions that are retrieved. Set all operators to ALL if you do not want to qualify based on hierarchy or trace field values. Qualifier values for both fields include:

<>: Greater than or less than the value entered.

=: Equal to the value entered.

ALL: All entries are returned, and there is no qualification on the field.

BETWEEN: Select the field between the values you enter.

LIKE: Select the field where it is like or similar to the value. Use the % character to denote wildcard positions.

Use the Perform Data Extraction page (QS_QSX_INQ) to execute the Data Extraction PeopleCode process.

Navigation:

Quality > Analyze Quality Results > Perform Data Extraction

The page displays the extraction criteria defined using the Define Extract Criteria component.

Field or Control

Description

Session History

If you haven't locked the qualifier selections that specify the streams and data content to extract from the database, you can change the values associated with the statements and the history. The new settings are in effect for this run only and aren't saved.

Search

Click this button to run the Data Extraction PeopleCode process. When the process is complete, the system creates a dataset based on the extraction contents. Using the browser, you can display the dataset online if the Quality application client is installed on a local machine, or you can download the file to a local machine.

Dataset Location

Displays the location of where the dataset is created upon completion of the extraction process.

Note: The Data Extraction process creates one file, in the location you specified on the File Locations page. The name of the file is the extraction name with this extension: cld.

View Dataset

Click this button to view the dataset using the browser. You can either download the dataset file to a local machine or open the dataset remotely using the Quality application client. In either case, it is assumed that you have installed the application client software on the computer that you are using.