While simulation and testing explore some of the possible behaviors and scenarios of the system, leaving open the question of whether the unexplored trajectories may contain the fatal bug, formal verification conducts an exhaustive exploration of all possible behaviors.

─Clarke, Grumberg & Peled

To me, mathematics, computer science, and the arts are insanely related. They’re all creative expressions.

─Sebastian Thrun

Haskell is faster than C++, more concise than Perl, more regular than Python, more flexible than Ruby, more typeful than C#, more robust than Java, and has absolutely nothing in common with PHP.

─Audrey Tang

This repository contains my work for obtain my computer scientist master degree: a simple model checker for temporal logics written in Haskell. My thesis is here. This work simplifies and improves Girish Bhat, Rance Cleaveland and Orna Grumberg work. See Efficient On-the-Fly Model Checking for CTL* for more details.

In summary, this repository contains a model verifier for LTL temporal logic, which, does not use Büchi's Automatons or Strongly Connected Components, is only based on formal semantics. This verifier extends naturally to the CTL* temporal logic (and thus, to the CTL logic).

• GHC 9.6.7 (The Glasgow Haskell Compiler)
• Haskell Cabal
• The new symbolic model checker nuXmv Version 2.1.0 (November 29, 2024) (don't forget edit the location path in the file Core.hs).

1. Installation:

cabal new-build

2. How to use it:

Let n the number of variables, and m the depth of the formulas of the experiments. Both n and m are positive integers.

• Run a random LTL (or CTL) experiment: cabal new-run model-check random LTL n m or cabal new-run model-check random CTL n m
• Run a random LTL (or CTL) experiment comparing with nuXmv: cabal new-run model-check random nuXmv LTL n m or cabal new-run model-check random nuXmv CTL n m
• Run a random LTL experiment with counterexamples: cabal new-run model-check random LTLc n m
• Run a random LTL experiment with counterexamples comparing with nuXmv: cabal new-run model-check random nuXmv LTLc n m
• Run a specific LTL experiment with counterexamples by using their initial seeds: cabal new-run -- model-check seeds ranInit ranNumInit ranKS ranF LTLc n m
• Run a specific LTL experiment with counterexamples by using their initial seeds comparing with nuXmv: cabal new-run -- model-check seeds nuXmv ranInit ranNumInit ranKS ranF LTLc n m
• Run a specific LTL or CTL experiment by using their initial seeds: cabal new-run -- model-check seeds ranInit ranNumInit ranKS ranF [LTL or CTL] n m
• Run a specific LTL or CTL experiment by using their initial seeds comparing with nuXmv: cabal new-run -- model-check seeds nuXmv ranInit ranNumInit ranKS ranF [LTL or CTL] n m
• Run simple LTL and CTL examples: cabal new-run model-check examples
• Run the experiments in my thesis: cabal new-run model-check thesis-experiments
• (NEW) Run an LTL experiment using parameterized formulas (m) over circular or random Kripke structures (n):

cabal new-run model-check LTL-experiment ks n experiment m (nuXmv)

n ≥ m, ks ∊ [cycleKS, randomKS],and experiment ∊ [fmTimesX, securityG, vivacidadF, progresoG, safeG, permGF, repeatG, altG, cicloCompletoG]