Today, many industrial production cells are controlled by software. Many such systems have to deal with requirements which the developer has to guarantee. Because of the complexity of the implementation one of the main problems for developing the software for reactive systems is to be sure that such properties are fulfilled. One way to handle the problems is to use formal methods: This means to develop a formal model which is used to prove the properties of the specification with tool support. There are many different methods to model such reactive systems. Some of these abstract from real-time aspects of the system. We chose a problem area where we have real-time requirements, for example the throughput of the modelled production cell. So we have to use formal methods which support models of real-time systems. In the past we looked for automata-based approaches. We used the concepts of timed and hybrid automata, for which there exist a well developed theoretical base ([AD94], [Hen96]) and some tools implementing these concepts as well as supporting reachability analysis. We used one of the tools, HyTech, for collecting experience in the domain of automatabased modelling techniques [Rus99]. For use in software development processes, this tool has some disadvantages, for example:
[1]
Rajeev Alur,et al.
A Theory of Timed Automata
,
1994,
Theor. Comput. Sci..
[2]
Martín Abadi,et al.
The existence of refinement mappings
,
1988,
[1988] Proceedings. Third Annual Information Symposium on Logic in Computer Science.
[3]
Dirk Beyer,et al.
Modeling a Production Cell as a Distributed Real-Time System with Cottbus Timed Automata
,
1998,
FBT.
[4]
Thomas A. Henzinger,et al.
The theory of hybrid automata
,
1996,
Proceedings 11th Annual IEEE Symposium on Logic in Computer Science.
[5]
N. V. Chernikoba.
Algorithm for discovering the set of all the solutions of a linear programming problem
,
1968
.