Tutorials

Scheduling and Personnel Rostering are two of the most successful application domains for Constraint Programming (CP). For many problems in these areas, CP provides the best compromise between a rapid development process, performance of the resulting tool, and the ability to keep the solution up-to-date with changing requirements. In this tutorial we focus on the modeling of such problems with different CP tools, how to identify and describe the constraints of a problem, and how to test and tune models for practical problem instances. We describe the main alternatives of describing such constraint models, either using high-level modeling languages, programming language APIs, or domain specific data models, and discuss the main advantages and disadvantages of each of these approaches. To illustrate these concepts, we present examples of a number of large scale, industrial applications in these domains, and give an overview of the existing literature in the field.

Over the last years, we have witnessed tremendous advances in the effective and efficient parallelization of crucial constraint solving frameworks, such as propositional satisfiability (SAT), satisfiability modulo theories (SMT), and mixed integer programming (MIP). Our tutorial addresses researchers, authors, and users of constraint solvers and provides an introduction and overview of cutting-edge parallel and distributed constraint solving methods and tools. Part I of the tutorial (D. Schreiber) introduces foundations of (massively) parallel computing, scheduling methods for irregular and unpredictable tasks, and evaluation methodology for parallel constraint solving. Part II (D. Schreiber) disseminates parallel propositional satisfiability (SAT), interspersed with cross-considerations and perspectives on parallel CP solving. Part III (S. Cai) provides an in-depth overview of parallelizations of SMT solving while Part IV (S. Cai) covers recent advances on parallel MIP solving.