Convex hull representations for bounded products of variables

It is well known that the convex hull of {(x, y, xy)}, where (x, y) is constrained to lie in a box, is given by the reformulation-linearization technique (RLT) constraints. Belotti et al. (Electron Notes Discrete Math 36:805-812, 2010) and Miller et al. (SIAG/OPT Views News 22(1):1-8, 2011) showed that if there are additional upper and/or lower bounds on the product z = xy, then the convex hull can be represented by adding an infinite family of inequalities, requiring a separation algorithm to implement. Nguyen et al. (Math Progr 169(2):377-415, 2018) derived convex hulls for {(x, y, z)} with bounds on z = xy(b), b >= 1. We focus on the case where b = 1 and show that the convex hull with either an upper bound or lower bound on the product is given by RLT constraints, the bound on z and a single second-order cone (SOC) constraint. With both upper and lower bounds on the product, the convex hull can be represented using no more than three SOC constraints, each applicable on a subset of (x, y) values. In addition to the convex hull characterizations, volumes of the convex hulls with either an upper or lower bound on z are calculated and compared to the relaxation that imposes only the RLT constraints. As an application of these volume results, we show how spatial branching can be applied to the product variable so as to minimize the sum of the volumes for the two resulting subproblems.

Automated summary

The study focuses on minimizing the volume of convex hulls with either upper or lower bounds on the product variable. It demonstrates that convex hulls with bounds on the product can be represented by RLT constraints, bounds on the product, and SOC constraints, with a maximum of three SOC constraints under certain conditions. Additionally, the volumes of convex hulls with bounds on the product are calculated and compared to a relaxation with only RLT constraints.