Developing Approaches for Solving a Telecommunications Feature Subscription Problem
This work addresses a specific telecommunications configuration problem, but it is incremental as it applies existing optimization methods to a new domain without major breakthroughs.
The paper tackled the NP-hard problem of finding optimal relaxations for inconsistent telecommunications feature subscriptions, which generalizes the feedback vertex set problem, by presenting and experimentally comparing constraint programming, partial weighted maximum Boolean satisfiability, and mixed integer linear programming formulations on randomly generated instances.
Call control features (e.g., call-divert, voice-mail) are primitive options to which users can subscribe off-line to personalise their service. The configuration of a feature subscription involves choosing and sequencing features from a catalogue and is subject to constraints that prevent undesirable feature interactions at run-time. When the subscription requested by a user is inconsistent, one problem is to find an optimal relaxation, which is a generalisation of the feedback vertex set problem on directed graphs, and thus it is an NP-hard task. We present several constraint programming formulations of the problem. We also present formulations using partial weighted maximum Boolean satisfiability and mixed integer linear programming. We study all these formulations by experimentally comparing them on a variety of randomly generated instances of the feature subscription problem.