|
Modeling
In many industries, model building is part of the project design and approval process. The model ensures that the pieces of the project fit together, and it gives a visual look to the design for the various stakeholders who will sign off on the project plan.
Centralized Master Data Management projects can
benefit from the modeling process in the same way.
Building an enterprise wide infrastructure around
corporate data involves understanding the business
information from both a high level view and a detailed
operational view. A data model can be reviewed against
the business processes and goals before the company
embarks on cross-departmental initiatives.
model :
- A
schematic description of a system, theory, or phenomenon that accounts for
its known or inferred properties and may be used for further study of its
characteristics.
A data model captures
information and data relationships in a design. The model centralizes the data descriptions stored in the repositories dispersed throughout your organization. The model is very data-centric, and accounts for all of the existing data types, codes, enumerations, and structures that are used within your schemas.
ObjectRiver MDM uses
your data model to generate a data dictionary for all to
review during the project design phase. When the project
plan is approved, the system uses the same model to
generate a transactional database infrastructure for the
enterprise.
The Modeling Process
The first step in creating a model
is to have your experts analyze your business objects
in a way that reflects the language used by business
users. For example, if you are modeling a health
care insurance system, you would expect your model
to contain objects like Claim, Policy,
Doctor, or Prescription
.
The system architect, data architect, and modeling
expert create the model for the new architecture. In
many organizations, it is the data architect who really
understands the intricacies of the existing data fields
and how they are used. All of this information must be
captured in the model.
Capturing the information in a model is usually a two
to eight week process, depending on the complexity of
the data. This exercise is totally focused on design; no
programming is involved. The iterative nature of the
review process lets you refine the model without having
to involve programmers.
Data Dictionary
The result of the modeling
process is a new data dictionary. The dictionary is
reviewed and validated by cross functional groups to
ensure that the business objects have the information
needed to perform the business processes. The data
dictionary can be printed in specification form and used
as a handbook for the programmers who are writing
business logic and rules for the new system.
This is really where the buy-off must occur,
because the data definition represents a long-term
strategy for data driven decision-making. You can
implement the data dictionary using the corporation's
relational database solution. In many cases, the new
data dictionary is very similar to many parts of the
existing schemas within the organization.
Generating
Business Objects
Once you have defined, modeled and reviewed the design of your business objects, the ObjectRiver Model Compiler can generate them from the same data model. The compiler generates database tables, Java data access source code, XML schemas for Web Services, and built-in support for auditing.
Process Objects
Process objects like
Online_Request, Work_Queue,
Exceptions , and the like can also be defined. These objects are more transient in nature, and may be handled by existing workflow systems.
Summary
The strategic modeling approach is an
inexpensive task that will prove that you can capture
and solve your data centralization problem. Modeling
your information essentially guarantees success because
once the model is defined, ObjectRiver can generate
the implementation for storing, distributing, and
auditing your business objects. Since the model can be
developed without programming, this process can be used
as a sales tool for the project manager, integrator or
workflow vendor trying to win business.

Business Rules
|