Ebook: High Integrity Software

The second half of the twentieth century has witnessed remarkable advances in technology. The unquestioned leader in this race has been computer technology. Even the most modest personal computers today have computing power that would have astounded the leading technol­ ogists a few decades earlier, and what's more, similar advances are pre­ dicted for many years to come. Looking towards the future, it has been conservatively estimated that in 2047 computers could easily be 100,000 times more powerful than they were in 1997 (Moore's law [Moore] would lead to an increase on the order of around 10 billion) [Bell]. Because of its enormous capability, computer technology is becoming pervasive across the technology spectrum. Nowadays it is not surpris­ ing to discover that very common household gadgets like your toaster contain computer technology. Televisions, microwave ovens, and even electric shavers contain software. And what's more, the use of computer technology has been estimated to double every two years [Gibbs]. In order to keep up with the growing technology demands and to fully utilize the ever more powerful computing platforms, software projects have become more and more ambitious. This has lead to software systems becoming dominant forces in system functionality. Further­ more, the ambition to realize significant portions of a system's function­ ality through software has extended into the high consequence realm. Presently, software controls many critical functions in (1) airplanes, (2) electronic commerce, (3) space-bound systems, (4) medical systems, and (5) various transportation systems such as automobiles and trains.

---

Within 50 years computers could have capabilities rivaling that of the human brain. Effective utilization of such new technologies poses a significant challenge to the computer science community, which finds an ever increasing number of complex applications within its technological grasp. In addition to increased complexity, most, if not all, of these applications are also accompanied by an inherent increase in the consequences associated with their failure, resulting in the construction of increasingly high consequence complex systems. Systems that fall within this domain are beyond the ability to construct in a brute force manner. There are two major challenges in developing such systems: manage complexity and provide sufficient evidence that the system satisfies dependability constraints.
Society is tacitly relying on the research community to solve these problems on a timetable satisfying the needs of industry. While impressive results have been obtained, the research community is still, to some extent, hamstrung by the lack of realistic case study problems against which to benchmark new techniques and approaches.
The purpose of High Integrity Software is to explore a cross-section of some of the most promising areas of research in the construction of high consequence complex systems, for example, a case study involving the Bay Area Rapid Transit (BART) system. Because of its scope and complexity, the BART case study is being recognized by many in the formal methods community as one of the definitive case study problems, and as such provides a valuable insight into the challenges that must be faced in the upcoming years.
High Integrity Software is suitable as a secondary text for a graduate level course, and as a reference for researchers and practitioners in industry.

---

Within 50 years computers could have capabilities rivaling that of the human brain. Effective utilization of such new technologies poses a significant challenge to the computer science community, which finds an ever increasing number of complex applications within its technological grasp. In addition to increased complexity, most, if not all, of these applications are also accompanied by an inherent increase in the consequences associated with their failure, resulting in the construction of increasingly high consequence complex systems. Systems that fall within this domain are beyond the ability to construct in a brute force manner. There are two major challenges in developing such systems: manage complexity and provide sufficient evidence that the system satisfies dependability constraints.
Society is tacitly relying on the research community to solve these problems on a timetable satisfying the needs of industry. While impressive results have been obtained, the research community is still, to some extent, hamstrung by the lack of realistic case study problems against which to benchmark new techniques and approaches.
The purpose of High Integrity Software is to explore a cross-section of some of the most promising areas of research in the construction of high consequence complex systems, for example, a case study involving the Bay Area Rapid Transit (BART) system. Because of its scope and complexity, the BART case study is being recognized by many in the formal methods community as one of the definitive case study problems, and as such provides a valuable insight into the challenges that must be faced in the upcoming years.
High Integrity Software is suitable as a secondary text for a graduate level course, and as a reference for researchers and practitioners in industry.
Content:
Front Matter....Pages i-xv
Front Matter....Pages 1-1
Designware: Software Development by Refinement....Pages 3-21
B: Towards Zero Defect Software....Pages 23-42
The Use of B to Specify, Design and Verify Hardware....Pages 43-62
A System for Predictable Component-Based Software Construction....Pages 63-87
Autonomous Decentralized Systems Technologies and Their Application to a Train Transport Operation System....Pages 89-111
Front Matter....Pages 113-113
Bay Area Rapid Transit District Advance Automated Train Control System Case Study Description....Pages 115-135
Using SCR to Specify Requirements of the Bart Advanced Automated Train Control System....Pages 137-168
On the Construction of a Domain Language for a Class of Reactive Systems....Pages 169-196
A Refinement-Based Approach to Deriving Train Controllers....Pages 197-240
Front Matter....Pages 240-240
Systematic Validation of a Relational Control Program for the Bay Area Rapid Transit System....Pages 243-264
Verification of a Controller for Bart: An Approach Based on Horn Logic and Denotational Semantics....Pages 265-299
Using Virtual Reality to Validate System Models....Pages 301-320
Back Matter....Pages 321-325

---

Within 50 years computers could have capabilities rivaling that of the human brain. Effective utilization of such new technologies poses a significant challenge to the computer science community, which finds an ever increasing number of complex applications within its technological grasp. In addition to increased complexity, most, if not all, of these applications are also accompanied by an inherent increase in the consequences associated with their failure, resulting in the construction of increasingly high consequence complex systems. Systems that fall within this domain are beyond the ability to construct in a brute force manner. There are two major challenges in developing such systems: manage complexity and provide sufficient evidence that the system satisfies dependability constraints.
Society is tacitly relying on the research community to solve these problems on a timetable satisfying the needs of industry. While impressive results have been obtained, the research community is still, to some extent, hamstrung by the lack of realistic case study problems against which to benchmark new techniques and approaches.
The purpose of High Integrity Software is to explore a cross-section of some of the most promising areas of research in the construction of high consequence complex systems, for example, a case study involving the Bay Area Rapid Transit (BART) system. Because of its scope and complexity, the BART case study is being recognized by many in the formal methods community as one of the definitive case study problems, and as such provides a valuable insight into the challenges that must be faced in the upcoming years.
High Integrity Software is suitable as a secondary text for a graduate level course, and as a reference for researchers and practitioners in industry.
Content:
Front Matter....Pages i-xv
Front Matter....Pages 1-1
Designware: Software Development by Refinement....Pages 3-21
B: Towards Zero Defect Software....Pages 23-42
The Use of B to Specify, Design and Verify Hardware....Pages 43-62
A System for Predictable Component-Based Software Construction....Pages 63-87
Autonomous Decentralized Systems Technologies and Their Application to a Train Transport Operation System....Pages 89-111
Front Matter....Pages 113-113
Bay Area Rapid Transit District Advance Automated Train Control System Case Study Description....Pages 115-135
Using SCR to Specify Requirements of the Bart Advanced Automated Train Control System....Pages 137-168
On the Construction of a Domain Language for a Class of Reactive Systems....Pages 169-196
A Refinement-Based Approach to Deriving Train Controllers....Pages 197-240
Front Matter....Pages 240-240
Systematic Validation of a Relational Control Program for the Bay Area Rapid Transit System....Pages 243-264
Verification of a Controller for Bart: An Approach Based on Horn Logic and Denotational Semantics....Pages 265-299
Using Virtual Reality to Validate System Models....Pages 301-320
Back Matter....Pages 321-325
....