NP-completeness of some optimal sequencing problems with a given grouping of elements

From MaRDI portal





For \(N=\{1,...,n\}\) and some subset \(M\subseteq N\) denote by c(M) the number of subsets C of M which have the form \(C=\{l,l+1,...,l+r\}\) with l-1\(\not\in M\) and \(l+r+1\not\in M\). For some permutation \(\pi =(i_ 1,...,i_ n)\) of the elements in N define \[ (1)\quad F(\pi)=\sum^{n}_{k=1}f(c(\{i_ 1,...,i_ k\})) \] where f is a nondecreasing function. Furthermore let \(N_ 1,...,N_ s\) be a partition of N into pairwise disjoint and nonempty sets \(N_ i\). Denote by \(\Pi\) the set of all permutations \(\pi\) of the form \(\pi =(\pi_{q_ 1},...,\pi_{q_ s})\) where \(\pi_{q_ i}\) is a permutation of the subset \(N_ i\) \((i=1,...,s)\). It is shown that the problem of finding a permutation \(\pi\in \Pi\) which minimizes (1) is NP-hard. However, if \(\Pi\) is restricted to the set of all permutations of the form \((\pi_ 1,...,\pi_ s)\) the problem can be solved in O(n log n) steps, which was shown in an earlier paper of the first author.











This page was built for publication: NP-completeness of some optimal sequencing problems with a given grouping of elements

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4723552)