A multi-objective optimization model of component selection in enterprise information system integration

•Couplings and cohesions.•IT architecture multi-objective optimization.•Sensitive analysis. Integrating legacy IT assets and new commercial software components together into a flexible IT architecture is one of open challenges facing modern enterprises today. Most of previous studies focused on usin...

Full description

Saved in:
Bibliographic Details
Published in:Computers & industrial engineering Vol. 115; pp. 278 - 289
Main Authors: Mu, Lifeng, Kwong, C.K.
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.01.2018
Subjects:
Component selection
Enterprise information technologies
Flexible IT architecture
Multiple objective programming
OR in software engineering
OR in software engineering
Flexible IT architecture
Component selection
Enterprise information technologies
Multiple objective programming
ISSN:0360-8352
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Couplings and cohesions.•IT architecture multi-objective optimization.•Sensitive analysis. Integrating legacy IT assets and new commercial software components together into a flexible IT architecture is one of open challenges facing modern enterprises today. Most of previous studies focused on using re-engineering to improve the flexibility of IT architectures, rather than employing optimization theory in architecture design problem, especially the problem of component selection and re-allocation in IT architecture. Moreover, a scant amount of literature is available on considering the architectural flexibility and integration cost simultaneously. To fill in this gap, based on a modified quantitative method of measuring the relationship between couplings and cohesions in architecture, we devise a nonlinear multi-objective binary integer programming to select components from legacy candidates and commercial candidates, and to group them into services under the service-oriented architecture (SOA) environment. The customized SPEA2 algorithm is further used to solve the problem, and some managerial insights are provided based on experiments and sensitivity analysis with the model.
ISSN:0360-8352
DOI:10.1016/j.cie.2017.11.013