[Paper Review] Theory of processes

Andrew M. Mironov|arXiv (Cornell University)|Sep 12, 2010

Formal Methods in Verification94 references3 citations

TL;DR

This paper presents a comprehensive theory of processes as a formal framework for modeling and verifying dynamic systems through algebraic operations, equivalences, and recursive definitions. It introduces key concepts like trace equivalence, strong equivalence, observational equivalence, and observational congruence, along with polynomial-time algorithms for equivalence recognition and process minimization, enabling rigorous analysis of concurrent and communicating systems, including those with message passing and composite operators.

ABSTRACT

The book gives a detailed exposition of basic concepts and results of a theory of processes. The presentation of theoretical concepts and results is accompanied with illustrations of their application to solving various problems of verification of processes. Along with well-known results there are presented author's results related to verification of processes with message passing, and there are given examples of an application of these results.

Motivation & Objective

• To establish a rigorous mathematical foundation for modeling the behavior of dynamic systems as processes.
• To develop formal operations (e.g., prefix, parallel composition, restriction) for constructing complex processes from basic actions.
• To define and analyze multiple equivalence relations—trace, strong, observational, and observational congruence—for comparing process behaviors.
• To provide algorithmic methods for recognizing strong and observational equivalence and minimizing processes to their canonical forms.
• To extend the theory to processes with message passing and composite operators, enabling modeling of distributed and concurrent systems.

Proposed method

• Defining a process as a labeled transition system with actions, states, and traces, using formal axioms for behavior representation.
• Introducing core operations: prefix action, empty process, alternative and parallel composition, restriction, and renaming to build complex processes.
• Establishing trace equivalence via observable action sequences and strong equivalence via bisimulation relations.
• Proposing a logical criterion and a polynomial-time algorithm (O(n²)) for recognizing strong equivalence using the relation μ(P₁, P₂).
• Defining observational equivalence and congruence based on behavioral indistinguishability under all contexts, with criteria using observational bisimulation and logical characterization.
• Extending the framework to processes with message passing using variables, expressions, functional symbols, and message-passing actions, and modeling them via flowcharts and composite operators.

Experimental results

Research questions

• RQ1 How can dynamic system behavior be formally modeled as a process using actions and state transitions?
• RQ2 What algebraic operations are needed to compose and manipulate processes systematically?
• RQ3 How can different notions of behavioral equivalence—trace, strong, observational, and observational congruence—be formally defined and distinguished?
• RQ4 What efficient algorithms exist for recognizing strong and observational equivalence between finite processes?
• RQ5 How can processes with message passing and composite operators be formally defined and verified for equivalence?

Key findings

• A polynomial-time algorithm (O(n²)) is presented for recognizing strong equivalence between finite processes using the relation μ(P₁, P₂).
• Minimization of finite processes with respect to strong equivalence is achieved via an algorithm that identifies and merges bisimilar states.
• Observational equivalence is characterized both logically and via observational bisimulation, enabling formal verification of behavioral indistinguishability.
• A method for proving observational equivalence of processes with composite operators is developed, supported by reduction and transformation techniques.
• The theory supports recursive definitions of processes through process expressions and provides a proof method based on recursive definitions and embeddings.
• The framework is extended to model message-passing systems using flowcharts and composite operators, with formal equivalence criteria and verification examples provided.