Challenges of Multi-Objective Optimization in Feature Model of Software Product Line

Abstract

Development of specific product with high reusability of existing resources, Software Product Line (SPL) is vastly used. Feature model is used to maintain the features of SPL in systematic way with common and variable features. Variable features differentiate the products of SPL domain. Optimization is required to find all possible combination of features for specific product under constraints and requirement of stakeholder. Multiple requirements and objectives from stakeholder make it hard to select best feature selection. Different multiobjective optimization algorithms are used in industry to get best optimized feature selection from feature model. In this paper we identify challenges and issues after systematic review of popular algorithms and their optimized results in large and small scale feature models.