Safety-critical systems, in the sense of software-based systems used in safety critical applications, are 'high-tech'. They are products of modern technology. Their effective, efficient and safe functioning depends not only on the devel opment of the right technologies but also on the right use of them. The safety of a system may be compromised not only by faults in the system but also by the use in the first place of an unreliable, unsafe, or unproved technology in its development. The key to the development and use of both technologies and systems is the human being. Until recently, the importance of human involvement, other than at the direct operational level, was hardly admitted. But now the unreliability of humans is recognised, as is the potential for latent faults to be introduced into systems at any point in their life cycles, by all who are in volved with them, including designers and strategic decision makers.

Safety-critical systems are found in almost every sector of industry. An understanding of the basis for the functioning of these systems is therefore vital to all involved in their operation. This book provides an overview of the whole subject

The programme for the Second Safety-critical Systems Symposium was planned to examine the various aspects of technology currently employed in the design of safety-critical systems, as well as to emphasise the importance of safety and risk management in their design and operation. assessment There is an even balance of contributions from academia and industry. Thus, industry is given the opportunity to express its views of the safety-critical domain and at the same time offered a glimpse of the technologies which are currently under development and which, if successful, will be available in the medium-term future. In the field of technology, a subject whose importance is increasingly being recognised is human factors, and there are papers on this from the University of Hertfordshire and Rolls-Royce. Increasingly, PLCs are being employed in safety-critical applications, and this domain is represented by contributions from Nuclear Electric and August Computers. Then there are papers on maintainability, Ada, reverse engineering, social issues, formal methods, and medical systems, all in the context of safety. And, of course, it is not possible to keep the 'new' technologies out of the safety-critical domain: there are papers on neural networks from the University of Exeter and knowledge-based systems from ERA Technology.

This book contains the Proceedings of the 6th Safety-critical Systems Sympo sium, the theme of which is Industrial Perspectives. In accordance with the theme, all of the chapters have been contributed by authors having an industrial af filiation. The first two chapters reflect half-day tutorials - Managing a Safety-critical System Development Project and Principles of Safety Management - held on the first day of the event, and the following 15 are contributed by the presenters of papers on the next two days. Following the tutorials, the chapters fa~l into five sub-themes - the session titles at the Symposium. In the first of these, on 'Software Development Tech nology', Trevor Cockram and others report on the industrial application of a requirements traceability model, Paul Bennett on configuration management in safety-critical systems, and Brian Wichmann on Ada. The next 5 chapters are on 'Safety Management'. In the safety domain, the fundamental business of management is increasingly being addressed with respect not merely to getting things done, but also to controlling the processes by which they are done, the risks involved, and the need not only to achieve safety but to demonstrate that it has been achieved. In this context, Gustaf Myhrman reveals recent developments for safer systems in the Swedish De fence, and Shoky Visram reports on the management of safety within a large and complex Air Traffic Control project.

Sensemaking in Safety Critical and Complex Situations: Human Factors and Design Human factors-based design that supports the strengths and weaknesses of humans are often missed during the concept and design of complex technical systems. With the focus on digitalization and automation, the human actor is often left out of the loop but needs to step in during safety-critical situations. This book describes how human factors and sensemaking can be used as part of the concept and design of safety critical systems in order to improve safety and resilience. This book discusses the challenges of automation and automated systems when humans are left out of the loop and then need to intervene when the situation calls for it. It covers human control and accepts that humans must handle the unexpected and describes methods to support this. It is based on recent accident analysis involving autonomous systems that move our understanding forward and supports a more modern view on human errors to improve safety in industries such as shipping and marine. The book is for human factors and ergonomists, safety engineers, designers involved in safety critical work and students. Stig Ole Johnsen is a Senior Researcher at SINTEF in Norway. He has a PhD from NTNU in Norway with a focus on resilience in complex socio-technical systems and has a Master’s in Technology Management from MIT/NTNU. He chairs the Human Factors in Control network (HFC) in Norway to strengthen the human factors focus during development and implementation of safety critical technology. His research interests include meaningful human control to support safety and resilience during automation and digitalization. Thomas Porathe has a degree in Information Design from Malardalen University in Sweden. He is currently Professor of Interaction Design at the Norwegian University of Science and Technology in Trondheim, Norway. He specializes in maritime human factors and design of maritime information systems, specifically directed towards control room design, e-navigation and autonomous ships. He has been working with e-Navigation since 2006 in EU projects such as BLAST, EfficienSea, MONALISA, ACCSEAS, SESAME and the unmanned ship project MUNIN. He is active in the International Association of Aids to Navigation and Lighthouse Authorities (IALA).

Today the vast majority of the world's information resides in, is derived from, and is exchanged among multiple automated systems. Critical decisions are made, and critical action is taken based on information from these systems. Therefore, the information must be accurate, correct, and timely, and be manipulated, stored, retrieved, and exchanged s

The contributions to this book are the invited papers presented at the fifth annual Safety-critical Systems Symposium. They cover a broad spectrum of issues affecting safety, from a philosophical appraisal to technology transfer, from requirements analysis to assessment, from formal methods to artificial intelligence and psychological aspects. They touch on a number of industry sectors, but are restricted to none, for the essence of the event is the transfer of lessons and technologies between sectors. All address practical issues and of fer useful information and advice. Contributions from industrial authors provide evidence of both safety con sciousness and safety professionalism in industry. Smith's on safety analysis in air traffic control and Rivett's on assessment in the automotive industry are informative on current practice; Frith's thoughtful paper on artificial intelli gence in safety-critical systems reflects an understanding of questions which need to be resolved; Tomlinson's, Alvery's and Canning's papers report on collaborative projects, the first on results which emphasise the importance of human factors in system development, the second on the development and trial of a comprehensive tool set, and the third on experience in achieving tech nology transfer - something which is crucial to increasing safety.