Book Image

Agile Model-Based Systems Engineering Cookbook

By : Dr. Bruce Powel Douglass
Book Image

Agile Model-Based Systems Engineering Cookbook

By: Dr. Bruce Powel Douglass

Overview of this book

Agile MBSE can help organizations manage constant change and uncertainty while continuously ensuring system correctness and meeting customers’ needs. But deploying it isn’t easy. Agile Model-Based Systems Engineering Cookbook is a little different from other MBSE books out there. This book focuses on workflows – or recipes, as the author calls them – that will help MBSE practitioners and team leaders address practical situations that are part of deploying MBSE as part of an agile development process across the enterprise. Written by Dr. Bruce Powel Douglass, a world-renowned expert in MBSE, this book will take you through important systems engineering workflows and show you how they can be performed effectively with an agile and model-based approach. You’ll start with the key concepts of agile methods for systems engineering, but we won’t linger on the theory for too long. Each of the recipes will take you through initiating a project, defining stakeholder needs, defining and analyzing system requirements, designing system architecture, performing model-based engineering trade studies, all the way to handling systems specifications off to downstream engineering. By the end of this MBSE book, you’ll have learned how to implement critical systems engineering workflows and create verifiably correct systems engineering models.
Table of Contents (8 chapters)

Chapter 2: System Specification

This chapter contains recipes related to capturing and analyzing requirements. The first four recipes are alternative ways to achieve essentially the same thing. Functional analysis generates high-quality requirements, use cases, and user stories, all of which are means to understand what the system must consist of.

By high-quality requirements, I mean requirements focused around a use case that are demonstrably the following:

  • Complete
  • Accurate
  • Correct
  • Consistent
  • Verifiable

The problem with textual requirements is that natural language is ambiguous, imprecise, and only weakly verifiable. Keeping text human-readable is very useful, especially for non-technical stakeholders, but is insufficient to ensure we are building the right system. The recipes covered in this chapter are as follows:

  • Functional analysis with scenarios
  • Functional analysis with activities
  • Functional analysis with state machines
  • Functional analysis with user stories
  • Model-based safety analysis
  • Model-based threat analysis
  • Specifying logical system interfaces
  • Creating the logical data schema