A Model for Hierarchical Open Real-Time Systems

Abstract

Introducing automated formal methods for large industrial real-time systems is an important research challenge. We propose timed process automata for modeling and analysis of time-critical systems which can be open, hierarchical, and dynamic. The model offers two essential features for large industrial systems: (i) compositional modeling with reusable designs for different contexts, and (ii) automated state-space reduction technique. Timed process automata model dynamic networks of continuous-time communicating control processes which can activate other processes. We show how to automatically establish safety and reachability properties of timed process automata by reduction to solving timed games. To mitigate the state-space explosion problem, an automated state-space reduction technique using compositional reasoning and aggressive abstractions is also proposed. Before working on timed process automata, we did a survey on semantics, decision problems, variants, and tools of timed automata. The insights gained from this survey motivated us to use timed game theory and Uppaal Tiga in a couple of industrial case studies and the development of timed process automata. Both the case studies show that state-space explosion is a severe problem for timed games. Suitable abstractions, however, dramatically improve the scalability of timed games in one case study. These case studies motivate the development of timed process automata and an automatable state-space reduction technique for them based on aggressive abstraction.