polyhedral combinatorics

polyhedral combinatorics
Name	Polyhedral combinatorics
Field	Mathematics
Related	Combinatorics, Discrete geometry

polyhedral combinatorics

Polyhedral combinatorics studies the combinatorial and geometric structure of convex polytopes and polyhedra, connecting discrete structures with continuous optimization and algebraic methods. It intersects with topics developed by mathematicians and institutions such as John von Neumann, Hugo Steinhaus, Paul Erdős, Hassler Whitney and research groups at Princeton University, Massachusetts Institute of Technology, University of Cambridge, Institut Henri Poincaré and Bell Labs. The field informs results used in work associated with awards like the Fields Medal, Abel Prize, Turing Award and organizations including the American Mathematical Society, Society for Industrial and Applied Mathematics and European Mathematical Society.

Definition and scope

Polyhedral combinatorics concerns the study of faces, vertices, edges and facets of polytopes linked to discrete problems introduced by researchers at University of California, Berkeley, Stanford University, University of Oxford and ETH Zurich; it leverages methods from contributors such as George Dantzig, Richard Karp, Jack Edmonds, László Lovász and Martin Grötschel. The scope includes links between canonical polyhedra arising in constructions studied by David Gale, Branko Grünbaum, Gian-Carlo Rota and facets of combinatorial optimization problems investigated at Bell Labs and IBM Research. Central objects often connect to classical results associated with Leonhard Euler, Augustin-Louis Cauchy and Eugène Charles Catalan as well as modern developments by scholars at California Institute of Technology, University of Illinois Urbana-Champaign and Columbia University.

Historical development

Foundations trace to classical geometry and to combinatorial advances by figures such as Leonhard Euler and later formalization by researchers at institutions like École Normale Supérieure and University of Göttingen. The modern field evolved through mid-20th century work by George Dantzig on linear programming, impactful combinatorial formulations from Jack Edmonds, and polytope enumeration studies by Branko Grünbaum and Gil Kalai. Important collaborative progress occurred in seminars and conferences organized by International Congress of Mathematicians participants, with further consolidation through monographs from presses such as Cambridge University Press, Springer-Verlag, Elsevier and lecture series at Institute for Advanced Study.

Key concepts and theorems

Fundamental concepts include convex hulls, face lattices, facet-defining inequalities, and unimodularity explored by mathematicians like Richard Stanley, Bernd Sturmfels, András Frank and Miklós Bóna. Central theorems and results link to the Four color theorem era techniques, the Hoffman–Kruskal theorem, the Birkhoff–von Neumann theorem, Farkas' lemma, and separation versus optimization paradigms advanced by Leonid Khachiyan and Rainer Karpinski. Structural results such as the g-theorem, developed in contexts related to work by Richard Stanley and Günter Ziegler, and polytope diameter bounds investigated by Václav Chvátal and László Lovász are pillars; enumerative combinatorics connections evoke contributions from Paul Erdős and Gian-Carlo Rota.

Polyhedral combinatorics in optimization

The discipline is central to integer programming, network flows, matching theory and cutting-plane methods originating from research by Gerard Cornuéjols, Jack Edmonds, Alexander Schrijver, George Dantzig and John Nash-adjacent optimization studies. Polyhedral descriptions of feasible regions underpin branch-and-cut and branch-and-bound frameworks developed at IBM Research, INRIA, Bell Labs and in industrial collaborations with companies such as AT&T and Ford Motor Company. Duality results and separation oracles are connected to algorithmic milestones like the ellipsoid method associated with Leonid Khachiyan and to complexity theory advances by Richard Karp, Stephen Cook and Arora-related probabilistic constructions.

Applications and examples

Polyhedral methods address the traveling salesman problem formulations explored by W. R. Hamilton-era successors and modern investigators like William Cook, the matching polytope studied by Jack Edmonds, and the stable set polytope analyzed by Claude Berge and later researchers at University of Grenoble and University of Paris-Sud. Other examples include formulations for scheduling problems relevant to work at General Electric and Siemens, facility location problems examined by academics at University of Toronto and Cornell University, and packing and covering instances appearing in collaborations with NASA and European Space Agency. Connections to algebraic geometry and toric varieties invoke scholars such as David Cox and Bernd Sturmfels while ties to statistical models reference contributions from Persi Diaconis.

Computational methods and software

Algorithmic and software advances include implementations of cutting-plane and polyhedral computations in systems developed at IBM Research, Microsoft Research, Zuse Institute Berlin and academic groups at INRIA and Max Planck Institute. Prominent software tools used by practitioners include solvers and libraries associated with projects at Gurobi-affiliated teams, COIN-OR initiatives, and academic packages influenced by work at Georgia Institute of Technology and University of Wisconsin–Madison. Computational benchmark collections and protocols are maintained by consortia like DIMACS and experimental evaluations appear in proceedings of conferences such as Symposium on Theory of Computing, International Symposium on Mathematical Programming and Conference on Integer Programming and Combinatorial Optimization.

Category:Mathematical disciplines