Relative Worst-Order Analysis: A Survey.

Joan Boyar, Lene M. Favrholdt, and Kim S. Larsen.
ACM Computing Surveys, 54(1):1-21, 2021. Article No. 8.

The standard measure for the quality of online algorithms is the competitive ratio. This measure is generally applicable, and for some problems it works well, but for others it fails to distinguish between algorithms that have very different performance. Thus, ever since its introduction, researchers have worked on improving the measure, defining variants, or defining measures based on other concepts to improve on the situation. Relative worst-order analysis (RWOA) is one of the most thoroughly tested such proposals. With RWOA, many separations of algorithms not obtainable with competitive analysis have been found.

In RWOA, two algorithms are compared directly, rather than indirectly as is done in competitive analysis, where both algorithms are compared separately to an optimal offline algorithm. If, up to permutations of the request sequences, one algorithm is always at least as good and sometimes better than another, the first algorithm is deemed the better algorithm by RWOA.

We survey the most important results obtained with this technique and compare it with other quality measures. The survey includes a quite complete set of references.

---

publication

Link to the publication at the publisher's site - subscription may be required.
Text required by the publisher (if any): The "open access" link below is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive version was published in ACM Computing Surveys, http://dx.doi.org/10.1145/3425910.

open access (321 KB)

The same as the publisher's version, when the publisher permits. Otherwise, the author's last version before the publisher's copyright; this is often exactly the same, but sometimes fonts, page numbers, figure numbers, etc. are different. It may also be a full version. However, it is safe to read this version, and at the same time cite the official version, as long as references to concrete locations, numbered theorems, etc. inside the article are avoided.

other publications

Other publications by the author.