Flexible design of operation behavior using modeling and visualization

Doctoral thesis, 2012

The behavior of a system can be described by a set of operations. The process of designing and developing these operations poses a real challenge in various situations, for example, when designing automation systems or keeping track of work in an emergency department. This thesis focuses on designing and developing the logical behavior of operations, for example, when they start, when they finish, and how they interact with the rest of the system. Being flexible during design and development entails coping with uncertainty. However, the practice of designing operation behavior is often fairly inflexible because every possible execution route is explicitly specified. This thesis suggests an operation design approach that does not explicitly specify operation routes. This is achieved in the proposed modeling method by only declaring current execution restriction using transition conditions in each operation, together with models representing entities, resources, and variables. A main benefit of the proposed modeling framework is the possibility it offers of creating multiple projections of operation relations to enable better understanding. This is accomplished by automatically generating a multiplicity of sequences showing various operation relations. The key discoveries that accomplish this are the state-based operation methods for relation identification and visualization. These methods make it possible to use the presented research not only when designing automation systems, but also in other applications, such as keeping track of all the patients in an emergency department. Some of the tools and methods presented here have been implemented in a prototype software called Sequence Planner.