Relationships: Improved Integrity for Relational-Based Computer Aided Design/Computer Aided Manufacturing/Geographic Information Systems

Relationships: Improved Integrity for Relational-Based Computer Aided
Design/Computer Aided Manufacturing/Geographic Information Systems

      Disclosed is a method for modeling complicated real world
relationships in Geographic Information Systems.  The solution is an
application of the entity-relationship model to the instance level of
a relational database system.  It provides robust insertion, update,
and deletion integrity beyond ordinary relational database
technology.  Four types of relationships are presented:
  o  one-to-one
  o  one-to-many
  o  many-to-one
  o  many-to-many

      The evolution of Geographic Information Systems (GIS), Computer
Aided Design (CAD), and Computer Aided Manufacturing (CAM) has
required that information be stored and retrieved from
non-proprietary databases  within an enterprise.  This information is
very complex.  In particular,  the relationships between real-world
objects modelled by GIS/CAD/CAM are  difficult to represent in most
relational databases.  These relationships  can be system defined or
user defined.  For a GIS, examples include:
  o  A house with a mailbox
  o  An ocean with ships and ports
  o  A port with ships and containers
  o  area features and their boundaries
  o  boundaries and their edges
  o  parent features and their dependent features
  o  layers and their features

      The trend for modeling these complex relationships has been the
adoption of object oriented databases and software by many
GIS/CAD/CAM systems.  Businesses have not benefitted from this
direction, however,  as most enterprises use relational databases.

      The problem with relational databases is that they do not
provide enough tools to model the complex relationships that exist in
the real world.  The major relational tools (referential integrity,
indirect tables and foreign keys) are too basic.  A GIS/CAD/CAM
system that forced users to be aware of these tools would be
extremely unwieldy.  What is needed is a unified approach to
modelling complex, real world relationships between instances of data
that uses an underlying relational database to preserve the
enterprise's investment.

      The name of the tool is relationships.  A relationship is an
interdependence between two instances of GIS/CAD/CAM data.  For
example, there could be a relationship between a telephone pole and a
road.  Relationships can only exist at the instance level; they
cannot exist between the "HEIGHT" attribute of a telephone pole and
the "WIDTH"  attribute of a road.

      Relationships can be system- or user-defined.  Examples of
system-defined relationships that are common to a GIS are:
  o  area features and their boundaries
  o  boundaries and their edges
  o  parent features and their dependent features
  o  layers and their features

Examples of user-defined relationships are:
  o  A house with a mailbox
  o  An ocean with ships and ports
  o ...