Remarks on: “Some Extensions of the Discrete Lotsizing and Scheduling Problem”

Abstract

Computational complexity results provide guideposts toward fruitful directions in algorithmic research and therefore play an important role in research on algorithm design. This note discusses complexity analysis in the context of lotsizing and scheduling problems. Such discussion is warranted for three reasons. First, research on problems that combine lotsizing and scheduling is growing rapidly. Second, these problems have the potential for requiring much less information to represent an instance than to represent a candidate solution. As a consequence, analysis may depend critically on subtle considerations relating to instance and solution size. Third, as we will see below, there is some evidence that the literature is unclear on this point.