This is the digital version of the printed book (Copyright © 2000).
Derek Hatley and Imtiaz Pirbhai—authors of Strategies for Real-Time System Specification—join with influential consultant Peter Hruschka to present a much anticipated update to their widely implemented Hatley/Pirbhai methods.
Process for System Architecture and Requirements Engineering introduces a new approach that is particularly useful for multidisciplinary system development: It applies equally well to all technologies and thereby provides a common language for developers in widely differing disciplines.
The Hatley-Pirbhai-Hruschka approach (H/H/P) has another important feature: the coexistence of the requirements and architecture methods and of the corresponding models they produce. These two models are kept separate, but the approach fully records their ongoing and changing interrelationships. This feature is missing from virtually all other system and software development methods and from CASE tools that only automate the requirements model.
System managers, system architects, system engineers, and managers and engineers in all of the diverse engineering technologies will benefit from this comprehensive, pragmatic text. In addition to its models of requirements and architecture and of the development process itself, the book uses in-depth case studies of a hospital monitoring system and of a multidisciplinary groundwater analysis system to illustrate the principles.
Compatibility Between the H/H/P Methods and the UML:
The Hatley/Pirbhai architecture and requirements methods—described in Strategies for Real-Time System Specification—have been widely used for almost two decades in system and software development. Now known as the Hatley/Hruschka/Pirbhai (H/H/P) methods, they have always been compatible with object-oriented software techniques, such as the UML, by defining architectural elements as classes, objects, messages, inheritance relationships, and so on. In Process for System Architecture and Requirements Engineering, that compatibility is made more specific through the addition of message diagrams, inheritance diagrams, and new notations that go with them. In addition, state charts, while never excluded, are now specifically included as a representation of sequential machines.
These additions make definition of the system/software boundary even more straightforward, while retaining the clear separation of requirements and design at the system levels that is a hallmark of the H/H/P methods—not shared by most OO techniques. Once the transition to software is made, the developer is free to continue using the H/H/P methods, or to use the UML or any other software-specific technique.
Table of Contents:
Foreword xvii
Preface xix
Figures and Tables xv
Part I: Concepts 1
Chapter 1: Introduction 3
1.1 Purpose and Scope 3
1.2 The System Development Process 4
1.3 Underlying Principles 4
1.2 What's in a Name? 5
1.3 Audience for and Structure of the Book 6
1.4 A Participative Case Study on the Web 7
1.5 A Caveat 8
Chapter 2: What Is a System? 9
2.1 System Characteristics 9
2.2 Views of a System 18
2.3 System Requirements 24
2.4 System Summary 40
Chapter 3: A Framework for Modeling Systems 41
3.1 A Model Framework 41
3.2 Models in General 41
3.3 Exploiting System Hierarchies 45
3.4 Exploiting the What/How Classification 49
3.5 Exploiting the Information/Material/Energy Classification 55
3.6 Layered Models: The Truth at Last! 60
3.7 Model Framework Summary 70
Chapter 4: System Development Models 72
4.1 Overview 72
4.2 Architecture Model 74
4.3 Requirements Model 112
4.4 Requirements Dictionary 162
4.5 Requirements/Architecture Relationships 170
4.6 A Note on Object Orientation 178
4.7 System Models Summary and Further Reading 180
Chapter 5: The System Development Process 181
5.1 Process, Methods, and Tools 181
5.2 The Nature of the Development Process 183
5.3 The Process and the Methods 190
5.4 Roles of the System Architect and System Engineer 200
5.5 System Development Process Summary 204
Chapter 6: Applying the Models to Development 205
6.1 Overview 205
6.2 Understanding the Generic Development Structure 206
6.3 Example: A Patient-Monitoring System 209
6.4 Configuring Software and Computer Hardware 236
6.5 Modeling the Numerous Hardware Technologies 244
6.6 Computer Hardware Layers 251
6.7 Software Layers 253
6.8 Summaries 273
Chapter 7: System Development Overview Using a Meta-Model 275
7.1 Introduction 275
7.2 A Meta-Model for Development Projects 276
7.3 An Essential Model of the Development Process 277
7.4 The Enhanced Development Model 281
7.5 The Development Architecture Context 284
7.6 Development Process Architecture 288
7.7 Development Process Task Allocation 291
7.8 Variations on the Architecture Template 291
Part II: Case Study: Groundwater Analysis System 297
Chapter 8: Initial Problem Statement 299
8.1 Overview 299
8.2 Required Capabilities 300
8.3 Required Performance 301
8.4 Required Constraints 301
Chapter 9: Modeling the Known Pieces 304
9.1 Overview 304
9.2 The Requirements Context 304
9.3 The System Timing Specification 305
9.4 The Entity Model 307
9.5 The Existing Sampling Module 311
Chapter 10: Building Upon the Known Pieces 316
10.1 Top-Level Essential Model 316
10.2 Enhancing the Essential Model 331
10.3 Architecture Context 341
10.4 Building Up from the Existing Sampling Module 343
10.5 What Do We Have, and What Is Missing? 346
Chapter 11: Filling in the Blanks 348
11.1 Introduction 348
11.2 Architecture Modules 348
11.3 Allocating the Enhanced Requirements Model 349
11.4 Enhancing the Allocated Models 357
11.5 Adding the Architecture Flows and Interconnects 362
11.6 Flow-to-Interconnect Allocations 362
11.7 Merging the Top-Down and Bottom-Up Pieces 364
Chapter 12: Completing the Models 375
12.1 Introduction 375
12.2 Accuracy Allocation 375
12.3 Timing Allocation–Concurrent Architecture Modules 376
12.4 Architecture Module Specifications 378
12.5 Architecture Interconnect Specifications 382
12.6 Requirements and Architecture Dictionaries 383
Chapter 13: Groundwater Analysis System Summary 387
13.1 Overview 387
Appendix: Changes, Improvements, and Misconceptions Since the Methods' Introduction 389
A. 1 A Learning Experience 389
A. 2 Changes and Improvements 390
A. 3 Misconceptions about the Methods 397
Glossary 401
Bibliography 419
Index 425
About the Author :
Derek J. Hatley is president emeritus of System Strategies, an international consulting and training firm based in Jenison, Michigan. He previously was a Staff Engineer at Smiths Industries, Avionic Systems Division, where he was responsible for systems and software development techniques. At Smiths, he developed the requirements model, managed its first major application, developed an instructional workshop to be given internally and to customers and suppliers, and presented the technique in technical papers at numerous international professional conferences. Retired for 12 years, he now enjoys a more relaxed life–with his wife of 52 years and his grown sons and their families, which include four grandchildren.
Peter Hruschka, a principal of The Atlantic Systems Guild, is based in Aachen, Germany. He is widely regarded as one of the fathers of CASE, and he helped create one of the first tools to implement the Hatley/Pirbhai real-time method. His mission: introducing better quality and higher productivity in system development as a trainer, consultant, and author. He has written more than a dozen books and numerous papers (mostly in German) on software architecture, requirements engineering, and development methods. Peter loves to travel and is often found with his wife, Monika, in some of the most scenic parts of the world, trying to hit little white balls into holes that are far too small.
Imtiaz A. Pirbhai was a principal of Systems Methods, based in Seattle, Washington. During his tenure at the Boeing Commercial Airplane Co., he was responsible for the development of structured methods standards and procedures. He specialized in real-time and general systems development training, utilizing the methods his team at Boeing introduced and developed. He taught and consulted internationally, introducing the methods described in Strategies for Real-Time System Specification (Dorset House, 1987) throughout the world. At the time of his death in 1992, he had begun work on the manuscript for Process for System Architecture and Requirements Engineering, completed by his coauthors Derek Hatley and Peter Hruschka.
