No abstract

This paper proposes a method for modelling and analysis of knowledge and commitments in multi-agent systems. The approach is based on an actors model and its reduction onto UPPAAL. A key factor of the approach is the possibility of exploiting the same UPPAAL model for exhaustive verification or, when state explosion problems forbid model checking, for quantitative evaluation of system properties through statistical model checking. The article describes the method, shows its application to modelling the NetBill protocol, proposes a translation into the terms of the timed automata language of UPPAAL and demonstrates the analysis of the NetBill protocol together with some experimental results.

This paper proposes a formal method based on the Theatre framework for modeling and analysis of knowledge and commitments in multi-agent systems (MASs). Theatre centers on actors and a reduction on to Uppaal, which enables both non-deterministic analysis (that is, qualitative analysis by exhaustive verification, or showing that something can possibly occur) and quantitative analysis (that is, estimating probability measures of event occurrence through simulations) of the same model. The article describes the modeling and analysis approach based on Theatre and Uppaal, and shows its application to the modeling and property checking of the NetBill protocol used in web-based MAS applications when selling/buying goods. Properties of the NetBill protocol are demonstrated by experimental results.

No abstract

No abstract