Publicación: El proceso de separación del empleado en una organización: aplicación de diagramas de influencia y cadenas de Markov en la toma de decisiones de recursos humanos
Cargando...
Fecha
2020-09-26
Autores
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
Atribución-NoComercial-SinDerivadas 4.0 Internacional
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial
Resumen
En toda organización existe un área de recursos humanos encargada, entre otras muchas cosas, de procesar toda la información relativa a los empleados, y con esa información realizar los procesos propios de esta área. Uno de los procesos más incómodos que, tarde o temprano se tiene que realizar, es cuando al empleado se le separa de su puesto, o bien es el propio empleado el que desea abandonarlo su puesto. El problema radica en que no siempre la organización desea que el empleado cause baja; lo quiere retener, pues el costo de esta baja a veces es incalculable, por lo que es necesario crear una herramienta que sirva como un indicador para anticipase a ese momento y si es posible revertirlo. El objetivo de este proyecto es ofrecer una propuesta de cómo construir esta herramienta para la toma de decisiones. Para crear esa herramienta se hará uso de diagramas de influencias, construyendo ese modelo a partir de la información con la que se gestiona al empleado. Entendiendo que el empleado pasa por diferentes estados en su desempeño y su valoración dentro de la organización hasta llegar a la baja. Estos estados se pueden modelar como un modelo de Markov que tiene que ser parte del diagrama de influencia que ayuda a determinar la situación en la que está el empleado. Esta herramienta se construye explotando y procesando la base de datos de recursos humanos para generar ficheros XML que puede ser usado por la aplicación OpenMarkov; para ello se hará uso de algoritmos implementados con lenguajes de programación. Cada uno de estos ficheros es una configuración diferente del mismo modelo, por lo cual hay que probar y evaluar hasta obtener el mejor de ellos. Una vez obtenido el mejor modelo, podemos ver que se puede aplicar para toma de decisiones en lo individual, para cada uno de los empleados, o en lo general, grupos enteros de empleados.
In every organization there is a human resources area in charge, among many other things, gathering and controlling all the information processes related to employees and, with that information carrying out the processes of this area. One of the most uncomfortable processes that, sooner or later, has to be implemented, is the employee’s separation from her or his position; sometimes it is the employee herself or himself who wishes to leave. The problem arises when the organization does not want the employee to leave but, on the contrary, it wants to retain him, because the cost of this leave is sometimes unmeasurable, ergo it is necessary to create a tool that serves as an indicator to anticipate that moment and, if possible, reverse it. The objective of this project is a proposal on how to build this tool for decision making. To create this tool, use will be made of influence diagrams, building that model using the information with which the employee is managed in the understanding that the employees go through different stages in their performance and their assessment within the organization until the employee’s leave arrives. These stages can be modeled as a Markov chain that has to be part of the Influence Diagram that helps determine the situation the employee is in. This tool is built by exploiting and processing the human resources database to generate XML files that can be used by the Open Markov application. For this, algorithms implemented with programming languages will be used. Each of these files is a different configuration of the same model, so we have to test and evaluate them until getting the best of them. Once the best model has been obtained, we can see that it can be applied to decisionmaking individually, for each of the employees, or in general, to entire groups of employees.
In every organization there is a human resources area in charge, among many other things, gathering and controlling all the information processes related to employees and, with that information carrying out the processes of this area. One of the most uncomfortable processes that, sooner or later, has to be implemented, is the employee’s separation from her or his position; sometimes it is the employee herself or himself who wishes to leave. The problem arises when the organization does not want the employee to leave but, on the contrary, it wants to retain him, because the cost of this leave is sometimes unmeasurable, ergo it is necessary to create a tool that serves as an indicator to anticipate that moment and, if possible, reverse it. The objective of this project is a proposal on how to build this tool for decision making. To create this tool, use will be made of influence diagrams, building that model using the information with which the employee is managed in the understanding that the employees go through different stages in their performance and their assessment within the organization until the employee’s leave arrives. These stages can be modeled as a Markov chain that has to be part of the Influence Diagram that helps determine the situation the employee is in. This tool is built by exploiting and processing the human resources database to generate XML files that can be used by the Open Markov application. For this, algorithms implemented with programming languages will be used. Each of these files is a different configuration of the same model, so we have to test and evaluate them until getting the best of them. Once the best model has been obtained, we can see that it can be applied to decisionmaking individually, for each of the employees, or in general, to entire groups of employees.
Descripción
Categorías UNESCO
Palabras clave
recursos humanos, separación del empleado, baja del empleado, abandono del puesto, cadenas de Markov, diagramas de influencia, employee dismissal, employee leave, abandonment of the position, Markov chains, influence diagrams
Citación
Centro
E.T.S. de Ingeniería Informática
Departamento
Inteligencia Artificial