Improving performance and scalability during product lines configuration in large variability domains

Product lines are used to model and derive a family of products that share a set of commonalities, while increasing customer satisfaction and decreasing time-to-market and production costs. Currently, the configuration of large product lines based on a set of non-functional properties such as costs and performance, is a problem of interest in the community. Semi-automatic configuration uses a set of configuration constraints such as optimization objectives (e.g. find the cheapest product) and hard limits (e.g. find a product that respects the budget) to guide the search. Techniques such as constraint programming and evolutionary algorithms are popular among the studied approaches. Nevertheless, state-of-the-art solutions have shown a performance and scalability issue when leading with large product lines, specially when using constraint programming. Moreover, the configuration of multiple product lines that are interrelated is a few explored field. Therefore, two research questions are defined in this document, mainly: RQ1. How to support the semi-automatic configuration based on constraint programming of multiple extended product lines by considering a set of configuration constraints?; and RQ2. What constraint programming heuristics can be used to improve performance and scalability when configuring multiple extended product lines, according to a set of configuration constraints?

• Autores: Ochoa Venegas, Lina María
• Fecha de publicación: 2017
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/13947
• Acceso en línea: http://hdl.handle.net/1992/13947
• Palabra clave: Ingeniería de software - Investigaciones; Programación con restricciones (Computación) - Investigaciones; Algoritmos heurísticos - Investigaciones; Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/