Further Results on an Abstract Model for Branching and its Application to Mixed-Integer Programming

TL;DR

This paper advances the theoretical understanding of variable selection in branch-and-bound algorithms for mixed-integer programming and demonstrates practical improvements in solver performance using the refined model.

Contribution

It solves open theoretical problems from previous work and implements improved model-based branching rules in a state-of-the-art MIP solver.

Findings

11% average reduction in solving time and nodes on benchmark instances

Improved theoretical understanding of variable selection in B&B

Enhanced performance of SCIP solver with new branching rules

Abstract

A key ingredient in branch and bound (B&B) solvers for mixed-integer programming (MIP) is the selection of branching variables since poor or arbitrary selection can affect the size of the resulting search trees by orders of magnitude. A recent article by Le Bodic and Nemhauser [Mathematical Programming, (2017)] investigated variable selection rules by developing a theoretical model of B&B trees from which they developed some new, effective scoring functions for MIP solvers. In their work, Le Bodic and Nemhauser left several open theoretical problems, solutions to which could guide the future design of variable selection rules. In this article, we first solve many of these open theoretical problems. We then implement an improved version of the model-based branching rules in SCIP 6.0, a state-of-the-art academic MIP solver, in which we observe an 11% geometric average time and node reduction on instances of the MIPLIB 2017 Benchmark Set that require large B&B trees.