Simulation of the software development process: an approximation using System Dynamics and the Larman Method

• Autores: Dugarte Peña, German Lenin, Sanchez Segura, Maria Isabel, Medina Domínguez, Fuensanta, de Amescua Seco, Antonio
• Formato: artículo
• Fecha de Publicación: 2020
• Institución: Universidad La Salle
• Repositorio: Revistas - Universidad La Salle
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.ulasalle.edu.pe:article/11
• Enlace del recurso: https://revistas.ulasalle.edu.pe/innosoft/article/view/11; https://doi.org/10.48168/innosoft.s1.a11; https://purl.org/42411/s1/a11; https://n2t.net/ark:/42411/s1/a11
• Nivel de acceso: acceso abierto
• Materia: Craig Larman's Method; Software Engineering Economics; Software Process Management; Software Process Simulation Modelling; System Dynamics; Dinámica de Sistemas; Gestión del Proceso Software; Método de Craig Larman; Modelado y Simulación de Procesos Software

The implementation of any software development process involves the consumption of critical resources. Software engineers cannot experiment with different development processes before starting them in real projects, due to the time that would entail and the amount of elements that are involved, so it is vital to have tools that allow the pre-visualization of the results of executing the software development process and how the environmental variables affect it, thus being able to anticipate under what conditions the software development process will be deployed. This paper presents the modelling and simulation of a software development process using System Dynamics (SD), which allows the graphical representation of the elements intervening in the software process, and the incorporation of as many relevant elements as possible. As a software costs estimation reference, the COCOMO estimation model was used; which beyond being reliable has a theoretical-practical foundation. As an ideal, and real, software process system, the Craig Larman Software Process model was chosen, also known as the Larman Method. The simulation model developed here, allows one to make some initial estimation of the software process and its elements’ behavior in the course of the simulation time. This is possible thanks to the observation and study of the system’s state variables, empowering one to discern about the effect of changes in the parameters on the general process, hence, carrying out relevant and interesting scenario studies. This model becomes a tool for supporting Software Project Management teams and enterprises whose business care on Technological Projects Management.