The Role of HazOp in Testing Health-Care Applications

Introduction The main challenge when testing safety-critical systems is to decide what to test. At the top level the answer to this question is straightforward – we should test that nothing dangerous can happen when we use the system. From this it follows that in order to test for safety we need to know and understand what can go wrong – we need to identify the system’s hazards. The proposed research paper will show how we can use HazOp to identify system hazards at an early stage in the development process. The identified hazards can be used to identify and build the test cases that can be used to ensure that the system will not be dangerous to put into operation. The results and experiences that will be reported stem from several sources – my own experiences as a consultant, work done together with a company developing software systems for the health services [1], research work and experiments done by two of my master students [2] and the ongoing research in the BUCS project [3]. HazOp is Useful in Testing Health Care Systems The two main worries for IT users in the health care are security – who has access to which information – and safety – can we trust the information we use for critical decisions. This work focuses on the safety part, not because we think security is not important but because our expertise and experience is mostly in the safety area. Lately, however, there has been a movement towards combining safety and security concerns and methods – see for instance [7]. To the extent that this holds, the methods proposed in this paper are also relevant for testing of security. Even though the focus in the proposed paper is on testing the system, the output from the HazOp should be used by all personnel that are involved with the system under consideration, such as: