ELECTRE TRI to classify items with respect to maintainability criterion

Risultato della ricerca: Other

Abstract

The paper aims to suggest a methodology able to support the analyst in the assessment process of items on the base of a criterion inspired to the maintainability parameter and in the successive assignment phase of the items in the predefined classes. The maintainability involves the continuous improvement of a system in order to improve the ability to maintain or improve the reliability of the analyzed system. In particular, the last one is the probability that a failed system will be restored to specified conditions within a given period of time when maintenance is performed according to prescribed procedures and resources. The faced case in this research regards the software maintenance, field in which the maintainability can be defined as the probability that a program will be restored to working conditions in a given period of time when it is being changed, modified or enhanced. Furthermore, the maintainability can be subdivided into diverse sub criteria as Analysability ( how easy or difficult is to diagnose the system for deficiencies or to identify the parts that need to be modified), Changeability (how easy or difficult is to make adaptations to the system), Stability (how easy or difficult is to keep the system in a consistent state during the modification process), Testability (how easy or difficult is it to test the system after modifications). Thus, with respect to aforementioned sub criteria the present research wants to propose the ELECTRE TRI method in order to assign the different analyzed items, with relation to the maintainability criterion, into the different predefined classes. With the aim to illustrate the method validity, a numerical application is shown with relation to the classification, under the maintainability criterion, of different modular Enterprise Resource Planning software systems.
Lingua originaleEnglish
Pagine1-5
Numero di pagine5
Stato di pubblicazionePublished - 2014

    Fingerprint

Cita questo