Publication: Método matemático de selección del rango de las variables de entrada en los modelos paramétricos de estimación software
Loading...
Identifiers
Publication date
2001
Defense date
2001-12-04
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
La estimación del software se refiere al conjunto de tareas encaminadas a determinar aspectos como el esfuerzo y el tiempo que senm necesarios para desarrollar un proyecto software. Entre las distintas metodologfas desarrolladas para llevar a cabo la estimación ocupa un lugar destacado la basada en modelos matematicos o paramétricos no lineales. Los modelos paramétricos utilizan un conjunto de variables independientes que a través de unas ecuaciones matemáticas proporcionan el valor de unas variables dependientes de salida. La correcta selección de los valores de entrada que se asignan a las variables independientes es un factor fundamental en el correcto funcionamiento de este tipo de modelos. El procedimiento de selecci6n de dichos valores utilizado por 10 principales modelos de estimación es actualmente cualitativo. EI principal objetivo de esta tesis es proporcionar un metodo matematico formal de selección de dichos parámetros de entrada que facilite la correcta selección de los mismos en cada proyecto concreto. El nuevo método se basa en la utilización de un conjunto de variables de entrada que se han denominado de segundo nivel y que permiten seleccionar el valor numerico adecuado de las variables que aparecen explicitamente en las ecuaciones y que se han denominado de primer nivel. Ademas se pretende que dicho metoda sea generico, esto es, adaptable a cualquier metoda matematico de estimación.
Software estimation refers to the set of tasks carried out in order to determine aspects such as effort and time needed to develop a software product. The mathematical or nonlineal parametric models stand out among the various methodologies developed to determine estimation With mathematical equations, parametric models use a set of independent variables to determine the value of the independent output variables. The correct selection of input values assigned to the independent variables IS a fundamental factor for the accurate estimation of this type of model. Currently, the selection procedure of the values used by the main estimation models is qualitative. The main objective of this thesis is to provide a formal mathematical selection method for these input parameters, which facilitates the correct selection of same for each specific project. The new method is based on the use of a set of input variables, called second level variable, and which permit the selection of the appropriate numerical value of the variables that explicitly appear in the equations and are called as first level. In addition, we hope that this method will be generic, that is, adaptable to any mathematical estimation model.
Software estimation refers to the set of tasks carried out in order to determine aspects such as effort and time needed to develop a software product. The mathematical or nonlineal parametric models stand out among the various methodologies developed to determine estimation With mathematical equations, parametric models use a set of independent variables to determine the value of the independent output variables. The correct selection of input values assigned to the independent variables IS a fundamental factor for the accurate estimation of this type of model. Currently, the selection procedure of the values used by the main estimation models is qualitative. The main objective of this thesis is to provide a formal mathematical selection method for these input parameters, which facilitates the correct selection of same for each specific project. The new method is based on the use of a set of input variables, called second level variable, and which permit the selection of the appropriate numerical value of the variables that explicitly appear in the equations and are called as first level. In addition, we hope that this method will be generic, that is, adaptable to any mathematical estimation model.
Description
Keywords
Diseño de software, Modelo matemático