Changes between Version 1 and Version 2 of ISO15926HowTo_Tutorial_RDSWIP_02

Show
Ignore:
Timestamp:
12/07/09 00:36:32 (14 years ago)
Author:
gordonrachar (IP: 198.53.220.181)
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • ISO15926HowTo_Tutorial_RDSWIP_02

    v1 v2  
    1313 
    1414---- 
    15 == Mapping to ISO 15926 Part 4 Using the RDS/WIP == 
    1615 
    17 Mapping databases together is the entry level for using ISO 15926.  The basic task is to examine each attribute to determine what it means, then find the appropriate class in the RDS/WIP, which is ISO 15926 part 4  (15926-4). 
    18  
    19 Benefits of using ISO 15926-4: 
    20  
    21   * It's already there.  (Creating a taxonomy of terms for something as complex as a petrochemical plant or refinery is not trivial.) 
    22   * It's there next time too.  (You don't have to keep reinvent it next time.  Re-use is easier every time you use it.) 
    23   * Other use it too.  (When you share information with a partner that has used ISO 15926-4 before, the infrastructure is already in place.  Both of you will have less adjusting to do.) 
    24  
    25 === Scenario === 
    26  
    27 In this scenario we will imagine an owner, an engineer, and a constructor exchanging information.  The owner wants early information about each pipe line to populate its line list, the engineer is designing the plant with a 3D design tool, and the constructor wants to feed the piping components to a purchasing application.  Each tool has its own database.  It is your job to map the information to common ISO 15926-4 classes. 
    28  
    29 Each participant requires two tables, one for pipe lines and one for valves. 
    30  
    31  
    32 === [wiki:ISO15926HowTo_Tutorial_DBscripts Create Database Scripts] === 
    33  
    34 The first step is to get a description of the database columns involved in the exchange.  In our example, we will assume that a database administrator in each of the three organizations gives you a SQL script that will create the tables at that organization, and another script to load a couple rows of sample data. 
    35  
    36 We prefer scripts that create the tables, rather than CSV file or spreadsheet, because the column names are sometimes ambiguous and the data types and sizes give valuable clues.  We want sample data for the same reason; when the column names are ambiguous, looking at some sample data will give a clue.  We prefer the data dumps in the form of scripts that will create the data rows because we invariably need to create and delete test data many times. 
    37  
    38 Assumption:  That you know enough of SQL databases to read the scripts and data dumps.  This will be necessary in the real world because you may have to build the database loading scripts from a printed data dump if the database cannot generate it for you. 
    39  
    40 You will only need to download one set of files.  We have provided the files in three forms, for Oracle, SQL Server, and SQL Server 2005 Express, because that is the systems the authors of this page used. 
    41  
    42 '''3D design system database''' 
    43   * attachment:3D_CreateTables_Ora.sql 
    44   * attachment:3D_LoadTables_Ora.sql 
    45   * attachment:3D_CreateTables_SS.sql 
    46   * attachment:3D_LoadTables_SS.sql 
    47   * attachment:3D_CreateTables_2K5.sql 
    48   * attachment:3D_LoadTables_2K5.sql 
    49  
    50 '''Owner Line List database''' 
    51   * attachment:LL_CreateTables_Ora.sql 
    52   * attachment:LL_LoadTables_Ora.sql 
    53   * attachment:LL_CreateTables_SS.sql 
    54   * attachment:LL_LoadTables_SS.sql 
    55   * attachment:LL_CreateTables_2K5.sql 
    56   * attachment:LL_LoadTables_2K5.sql 
    57  
    58 '''Constructor Material Control database''' 
    59   * attachment:MM_CreateTables_Ora.sql 
    60   * attachment:MM_LoadTables_Ora.sql 
    61   * attachment:MM_CreateTables_SS.sql 
    62   * attachment:MM_LoadTables_SS.sql 
    63   * attachment:MM_CreateTables_2K5.sql 
    64   * attachment:MM_LoadTables_2K5.sql 
    65  
    66 === [wiki:ISO15926HowTo_Tutorial_Spreadsheet Creating a Mapping Spreadsheet] === 
    67  
    68 The first step is to look at the three sets of data and create a cross-reference list for each table.  We want to be able to show equivalent database rows in the same row of the list.  A spreadsheet is an excellent medium. 
    69  
    70 Create a pair of columns for each participant, and three columns for ISO 15926-4 information: 
    71  
    72   * 3D Table 
    73   * 3D Column 
    74   * Line List Table 
    75   * Line List Column 
    76   * Material Table 
    77   * Material Column 
    78   * ISO 15926 Class Label 
    79   * ISO 15926 Class ID 
    80   * ISO 15926 Description 
    81  
    82 Sample Spreadsheet 
    83   * attachment:MappingSpreadsheet_Blank.xls 
    84  
    85 Examine the columns in each database listing and arrange equivalent column names on the same row.  The example below starts with the columns in the 3D table, with the Line List columns next, followed by the Material columns. 
    86  
    87 Sample Spreadsheet 
    88   * attachment:MappingSpreadsheet_Columns.xls 
    89  
    90 The first thing you notice is that there is not an exact match between the column names, and that each database contains information that neither of the other two does.  This is OK.  With ISO 15926 we do not have to share all of our information with our business partners, only that which has value. 
    91  
    92  
    93 === [wiki:ISO15926HowTo_Tutorial_RDSWIP_01 Mapping With the RDS/WIP] === 
    94  
    95 The RDS/WIP is at this website: 
    96  
    97   * rdl.rdlfacade.org 
    98  
    99 For basic instructions in how to use the RDS/WIP: 
    100  
    101   * [wiki:ISO15926HowTo_15926Tools_RDS_WIP Using the RDS/WIP (RDL Façade)] 
    102  
    103 Basically, your task here is to search for ISO 15926-4 class that most closely matches the meaning of each of the column names in your three databases.  The first thing you will notice is that the best choice is not immediately apparent.  In the sample spreadsheet below, some of the data rows have more than one choice.  Look at a few of them to see what the differences are. 
    104  
    105 Sample Spreadsheet 
    106   * attachment:MappingSpreadsheet_First.xls 
    107  
    108 === [wiki:ISO15926HowTo_DiscoverTheRightClass Discovering the Right Class] === 
    109  
    110 Here are some guidelines for selecting the best class: 
    111  
    112    
    113 The sample spreadsheet below has had the duplicates removed, with the best choice left. 
    114  
    115 Sample Spreadsheet 
    116   * attachment:MappingSpreadsheet_Second.xls 
    117  
    118 === [wiki:ISO15926HowTo_Tutorial_RDSWIP_02 Resolving Differences] === 
     16== [wiki:ISO15926HowTo_Tutorial_RDSWIP_02 Resolving Differences] == 
    11917 
    12018 
     
    12422 
    12523----  
    126 == [wiki:ISO15926HowTo_Tutorial_Templates Introduction to Templates] == 
    127  
    128 ... 
    129 ... 
    130 ... 
    131  
    132 ---- 
    133 == Something Else == 
    134  
    135 ... 
    136 ... 
    137 ... 
    138  
    139  
    140 ---- 
    141  
    14224== Back == 
    14325 
Home
About PCA
Reference Data Services
Projects
Workgroups