Closing the gap for single resource constraint scheduling

TL;DR

This paper introduces an improved scheduling algorithm for single resource constraint problems, achieving an approximation ratio close to the theoretical limit, thus nearly closing the gap between feasible solutions and computational hardness.

Contribution

It presents a new algorithm with an approximation ratio of (3/2 + ε), significantly improving upon the previous ratio of 2 + ε, and nearly matching the inapproximability bound.

Findings

New algorithm achieves (3/2 + ε) approximation ratio.

Closes the gap between inapproximability and existing algorithms.

Advances the state-of-the-art in resource-constrained scheduling.

Abstract

In the problem called single resource constraint scheduling, we are given $m$ identical machines and a set of jobs, each needing one machine to be processed as well as a share of a limited renewable resource $R$. A schedule of these jobs is feasible if, at each point in the schedule, the number of machines and resources required by jobs processed at this time is not exceeded. It is NP-hard to approximate this problem with a ratio better than $3/2$. On the other hand, the best algorithm so far has an absolute approximation ratio of $2+\varepsilon$. This paper presents an algorithm with absolute approximation ratio~$(3/2+\varepsilon)$, which closes the gap between inapproximability and best algorithm except for a negligible small~$\varepsilon$.