Business Rule Based Configuration Management and Software System Implementation Using Decision Table

1
Business Rule Based Configuration Management and
Software System Implementation Using Decision
Tables

• Olegas Vasilecas, Aidas Smaizys
• VGTU, Vilnius, Lithuania
• ADBIS 2007
• Presented by Aidas Smaizys

2
Introduction

• In this presentation we purpose to discuss
• Deployment and customization of the software in
  different information systems (IS) of separate
  organizations.
• Business rule (BR) based Configuration management
  and use for IS engineering.
• Automated BR based configuration implementation
  into the Software system (SS) code.
• As the possible solution we will present a method
  and the results of current experimental
  evaluation made.

3
Overview of the problem domain

• BRs of the same problem domain even in similar
  organizations are similar
• Similar legal regulation
• Similar business processes
• Similar infrastructure
• Similar integration needs
• Similar experience of the employees

• Why BRs are never applied the same way?
• Different business policy
• Different business processes
• Different infrastructure
• Different system integration needs
• Different experience of the employees

4
Problem formulation

• Existing system engineering methods are not
  capable enough of dealing with the complex and
  dynamic nature of Business rules in complex,
  dynamic and slightly different business
  environment of every specific business system in
  separate organizations.

5
Existing solution of the problem 1

• Splitting the source code line into the
  alternative paths, implementing specific business
  requirements by redesigning models through all
  the entire layers (business system, information
  system and software system layer) and producing
  separate executable code

6
Existing solution of the problem 2

• Design and implementation of modular software
  system architecture and customization of the
  separate modules according to the customers needs

7
Existing solution of the problem 3

• Implementation of all the possible variations of
  the same business rules application into the
  software code and activation using specific
  parameter values for every particular
  implementation

8
Existing solution of the problem 4

• Business system adaptation according to the
  constraints caused by particular implementation
  of business rules hard coded into the software
  system.

9
The proposed solution

• The method for automated business rule based
  configuration implementation and configuration
  management into the information system for use by
  software systems and services.

10
The method
11
The method. Main engineering processes

• Decision table (DT) transformation into the ER
  model and generation of DBMS for initial
  configuration.
• ER model redesign and accommodation for Business
  level and Software system (SS) code generation.

12
1. Engineering context vs. Architectural
contextDT transformation into the DBMS
13
DT and its metamodel
14
ER and DT metamodel (MOF/UML) mapping
15
Transformation model (XSLT)
16
Example of Configuration representation in DT
17
Example of DT transformation into the ER model
and generated tables in the DBMS with initially
filled values
18
2. Engineering context vs. Architectural
contextBusiness level ER based SS code
generation
19
Example of generated source codePass issue
process interface.
20
Experimental evaluation

• The method was used for development of automated
  security system in the sea port
• Transport and visitor control subsystem
• Subsystem for interactive data exchange with
  security system controllers
• Subsystem for logical control code generation
  (Horn clauses) and upload into the security
  system controllers
• Subsystem for data exchange with external servers

21
Contribution

• The proposed method has introduced an idea of
  using BR represented in DTs for IS development,
  customization and evolution
• BR based SS adaptation to the particular business
  system needs.
• Metamodel based transformations for automated
  implementation of BRs, formally represented in
  DTs and changing all the dependent models
  through the layers down until the implementation
  into the SS configuration (DBMS and SS code).

22
Conclusions

• The proposed method allowed automated
  implementation of BR into the SS configuration
• configuration parameter semantics can be
  expressed for business users in plain text
• Configuration parameter sets (collections) can be
  designed, viewed and edited using DTs or ERs by
  system analysts and designers
• Programmers do not influence BR based part of the
  SS code it is made automated way
• The proposed method allowed customization and
  deployment of the software in different
  information systems of separate organizations
• by changing configuration parameter values
• by running semi automated engineering process
  using existing transformations.

23
Thank you!

• Your questions please?