On the morphological structure of a network

Abstract

In literature, both morphological and resource-related measures are used to predict the difficulty of a project scheduling problem. Rapid progress regarding solution procedures has resulted in the development of a number of data generators in order to generate instances under a controlled design and in different standard sets with problem instances. These complexity measures need to serve as predictors for the complexity of the problem under study. In this paper, we report on results for the morphological structure of a network. The contribution of this paper is threefold. First, we review the existing literature of complexity measures and their link with the existing network generators up-to-date. Second, we review six morphological network indicators in order to describe each network structure in a detailed way. These indicators were originally developed by [27] and have been modified or sometimes completely replaced by alternative indicators in order to give a better description of the morphology of a network. Last, we generate a large amount of different networks with four network generators. This allows us to draw conclusions on both the performance of different network generators and the usefulness of the indicators measuring the morphological structure of a network and to give a critical remark on well-known datasets from literature. Our general conclusions are that none of the network generators are able to capture the complete feasible domain of all networks with a given input parameter. Moreover, each network generator covers its own network-specific domain and, consequently, contributes to the generation of instance data sets.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 26 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.