An Asymptotic $\left(\frac{4}{3}+\varepsilon\right)$-Approximation for the 2-Dimensional Vector Bin Packing Problem

We study the $2$-Dimensional Vector Bin Packing Problem (2VBP), a generalization of classic Bin Packing that is widely applicable in resource allocation and scheduling. In 2VBP we are given a set of items, where each item is associated with a two-dimensional volume vector. The objective is to partition the items into a minimal number of subsets (bins), such that the total volume of items in each subset is at most $1$ in each dimension. We give an asymptotic $\left(\frac{4}{3}+\varepsilon\right)$-approximation for the problem, thus improving upon the best known asymptotic ratio of $\left(1+\ln \frac{3}{2}+\varepsilon\right)\approx 1.406$ due to Bansal, Elias and Khan (SODA 2016). Our algorithm applies a novel Round&Round approach which iteratively solves a configuration LP relaxation for the residual instance and samples a small number of configurations based on the solution for the configuration LP. For the analysis we derive an iteration-dependent upper bound on the solution size for the configuration LP, which holds with high probability. To facilitate the analysis, we introduce key structural properties of 2VBP instances, leveraging the recent fractional grouping technique of Fairstein et al. (ESA 2021).