Computation of the \(\omega\)-primality and asymptotic \(\omega\)-primality with applications to numerical semigroups. (Q2339617)

There are many non-unique factorization invariants of finitely generated atomic monoids that have been the subject of extensive investigation in recent years. This paper focuses on the \(\omega\)-invariant, which in some sense measures how far an element is from being prime. Let \(S\) be an atomic monoid. For \(s\) a nonunit in \(S\), we define \(\omega(x)=n\) if \(n\) is the smallest positive integer with the property that whenever \(x\) divides a product of \(t\) atoms of \(S\), there is a collection of at most \(n\) of these atoms so that \(x\) divides their product. This is meant to extend the common definition that \(p\) is prime if and only if \(p\mid ab\) implies \(p\mid a\) or \(p\mid b\). The authors give an algorithm to compute the \(\omega\)-invariant of any element in any finitely generated atomic monoid. They use the fact that such a monoid is isomorphic to \(\mathbb N^p/\sigma\) for some congruence \(\sigma\) and an earlier result of \textit{V. Blanco, P. A. García-Sánchez}, and \textit{A. Geroldinger}, showing that \(\omega(x)\) is given by the maximum factorization length of a minimal element in a certain ideal [Ill. J. Math. 55, No. 4, 1385--1414 (2011; Zbl 1279.20072)]. An algorithm to compute the complete set of minimal elements of such an ideal is given in the paper of \textit{J. C. Rosales, P. A. García-Sánchez}, and \textit{J. I. García-García} [J. Algebra 238, No. 1, 328--344 (2001; Zbl 1001.20050)]. The authors give runtime information comparing this algorithm to another algorithm in the particular case of numerical semigroups. The authors also carefully analyze cancellative reduced monoids minimally generated by two elements. In this case, it is not difficulty to give a precise characterization of the congruence \(\sigma\) determining this monoid. The authors use this to give a formula for the \(\omega\)-primality of any element of such a monoid. This result nicely generalizes one of the main results of \textit{D. F. Anderson, S. T. Chapman, N. Kaplan}, and \textit{D. Torkornoo} [Semigroup Forum 82, No. 1, 96--108 (2011; Zbl 1218.20038)], which applies only to numerical monoids. The final and most difficult section of this paper is dedicated to computing the asymptotic \(\omega\)-primality in quasi-Archimedean cancellative reduced monoids. Let \(\overline\omega(x) = \lim_{n\to\infty}\frac{\omega(nx)}{n}\). The authors compute this asymptotic \(\omega\)-primality for all elements in a large class of monoids and highlight the application to numerical semigroups.