Modeling Software Development Methodologies: A Logic Based Approach

TL;DR

This paper proposes a logic-based approach to model and analyze software development life-cycles, using semantic web and rule-based languages to improve project management and tool support.

Contribution

It introduces encoding of software development life-cycles as workflows in semantic web and rule-based languages, demonstrating the use of Transaction Logic for agent-oriented methodologies.

Findings

Transaction Logic effectively models software development life-cycle.

The approach enables analysis of life-cycle properties.

CASE tools can utilize the logic-based models for guidance.

Abstract

In the last two decades, the growing trend of software development industry has made different aspects of software engineering more interesting for the computer science research community. Software development life-cycle is one of these aspects that has a significant impact on the success/failure of software development projects. Since each software development methodology relatively provides its own software development life-cycle, encoding software development life-cycle in a workflow representation language can help developers to properly manage their software projects. In addition, such encoding can also be used in CASE tools. In this report, we consider the software development life-cycle as a workflow that can be represented by semantic web and rule based languages. Such consideration let one analyze the properties of the life-cycle. Specifically, we take a well-known agent oriented software development methodology and show that its corresponding life-cycle can be specified by Transaction Logic easily and concisely. Finally, the compact and clear representation of the life-cycle in Transaction Logic can be used in CASE tools to guide software developers.