BSR lecture: Thursday July 5th, 11:00-12:30
Title: A Modest Tutorial in Quantitative Modelling and Analysis
Abstract: Over the last three decades, significant progress has been made in the area of formal methods to allow the construction and analysis of mathematically precise models of critical systems. Classically, model checking has been used to verify functional properties related to correctness and safety. However, since e.g. correct system implementations may still be unusably slow, performance requirements have to be considered as well. This need to evaluate both qualitative as well as quantitative properties fostered the development of integrative approaches that combine probabilities, real-time aspects and costs with formal verification techniques.
Today, behavioural modelling and analysis is supported by a wide range of tools and formalisms. In this lecture, we give a practical introduction to quantitative verification using the Modest modelling language. Modest is rooted in process algebra, but borrows syntax and concepts from widely-used programming languages to be more accessible to programmers and engineers. We present its mathematical foundations, ranging from Markov chains to priced probabilistic timed automata. The analysis of Modest models is supported by the Modest Toolset with exhaustive and statistical model checking techniques. Our lecture concludes with a hands-on experience to model and optimise the experiment scheduling on nanosatellites and attacks on the Bitcoin cryptocurrency.
Short bio: Jaco van de Pol studied Computer Science at Utrecht University, and received his PhD from the Faculty of Philosophy in Utrecht (1996) on Termination of Higher-order Rewrite Systems. After positions at the LMU Munich, TU/Eindhoven and CWI Amsterdam, he became full professor in Formal Methods and Tools at the University of Twente, being Head of the CS Department from 2014-2017. He was tool chair of TACAS 2015, and he is in the editorial board of the journals SCP and STTT. He was invited professor at the lab LIPN of Université Paris 13, both in 2016 and in 2017. His research interests include symbolic methods and parallel algorithms for verification and testing. Recent achievements are in scalable multi-core NDFS and multi-core SCC algorithms, and in multi-core Decision Diagram data-structures. These high-performance and/or symbolic algorithms are implemented in the LTSmin toolset, which won several prizes. The algorithms are available through a high-level API to multiple modelling formalisms, like Timed Automata (Uppaal), Promela (SPIN), Petri-Nets (PNML), Process Algebras (mCRL2), DiVinE, and B CSP (ProB). He applied verification technology in projects on railway interlockings, energy aware scheduling, biological signalling networks, and socio-technical security models.
Arnd Hartmanns is an assistant professor in the Formal Methods and Tools group at the University of Twente. His primary research interests are modelling tools and formalisms for stochastic timed and hybrid systems (in particular Modest) and their applications in various fields. Arnd was previously a postdoc in the Formal Methods and Tools group at the University of Twente and the Dependable Systems and Software group at Saarland University, where he also completed his Ph.D. in computer science with a thesis On the Analysis of Stochastic Timed Systems in 2015.