Software component evaluation and selection using TOPSIS and fuzzy interactive approach under multiple applications development

In this paper, a two phase approach is proposed for decision-making situation that involves optimal software component evaluation and selection for designing component-based modular software system with multiple applications. In the first phase, components are evaluated using technique for order preference by similarity to ideal solution. In the second phase, a non-linear multi-objective optimization model is developed that facilitates the decision whether to buy commercial-off-the-shelf components or to build in-house components so that the total score of alternative components is maximized while the overall cost and delivery time of the system are minimized. Many critical parameters such as reliability of various applications, reusability and compatibility of the software components are considered simultaneously in the proposed optimization model. To determine a preferred compromise solution for the multi-objective optimization problem, a fuzzy interactive approach is used. Numerical illustrations based on a small-scale case study are presented to demonstrate usefulness of the proposed optimization model for optimal build or buy decisions in real-world applications.

Automated summary

This paper proposes a two-phase approach for decision-making in designing component-based modular software system with multiple applications. The first phase evaluates components using technique for order preference by similarity to ideal solution. The second phase develops a non-linear multi-objective optimization model to determine whether to buy commercial-off-the-shelf components or to build in-house components. Critical parameters such as reliability, reusability, and compatibility are considered in the proposed model.