Estimación del Coste de la Calidad del Software a través de la Simulación del Proceso de Desarrollo

Some of the main objectives of quality assurance practices are to detect and correct the errors that arise during the development of a software product, as well as to guarantee that the final product conforms to the client’s requirements. Quality initiatives suppose an added cost to the software development projects, cost that is totally justified but which, in most of the cases, is the first one that is cut when the pressure of the budget and/or the time unexpectedly increases. In many occasions, the cost estimation of the quality initiatives consists of a percentage of part of the total budget considered for project development. This way, it is impossible to determine which the efficiency of those activities has been and how the quality problems arisen during the development could have been improved, if not solved. In this work we contribute a new approach to measure the efficiency, in the scope of cost and yield, of the software assurance activities. 1 Dpto. de Lenguajes y Sistemas Informáticos. Universidad de Cádiz. Escuela Superior de Ingeniería. C/ Chile, no1 11003 – Cádiz (Spain) e-mail: [email protected] 2 Dpto. de Lenguajes y Sistemas Informáticos. Universidad de Sevilla Facultad de Informática y Estadística. Avda. Reina Mercedes, s/n 41012 – Sevilla (Spain) e-mail: [email protected] 2 Mercedes Ruiz Carreira / Isabel Ramos Román r i rr ir / I l Tabla 1: Definición de las categorías de costes de la calidad Mientras que los costes de las actividades de aseguramiento de calidad y de la mejora del proceso han sido objeto de mucha investigación durante más de 20 años [Alberts 1976], no hay una progresión similar en la determinación del coste de la calidad del software (CoSQ) [Hersh 1993] y [Knox 1993]. El modelo de Knox y los estudios de Raytheon [Dion 1993] constituyen notables excepciones ya que sí emplean de forma explícita el coste de la calidad del software. Los trabajos de Knox We propose to combine the simulation of dynamic models, as an alternative to the expensive, and in most of the occasions impossible, experimentation in the project management field, with the CoSQ method.