Tighter quadratically constrained convex reformulations for semi-continuous quadratic programming

Abstract

The paper proposes a novel class of quadratically constrained convex reformulations (QCCR) for semi-continuous quadratic programming. We first propose the class of QCCR for the studied problem. Next, we discuss how to polynomially find the best reformulation corresponding with the tightest continuous bound within this class. The properties of the proposed QCCR are then studied. Finally, preliminary computational experiments are conducted to illustrate the effectiveness of the proposed approach.