Nowadays, change is an important aspect of the world. Complexity and change in requirements and environments bring us to Autonomic Systems as a solution. Like any other kind of software system better implementation of a system needs a proper evaluation method for system implementation. To evaluate a Self-Adaptive system, appropriate quality factors are needed for evaluation. This research tries to use non-adaptive systems evaluation methods to evaluate Self-Adaptive ones. The qualitative factors for self-adaptive systems have been extracted from a literature review (as Self-Adaptive System Qualitative Factors or for abbreviation SAQFs). Hence, there is no explicit or even implicit way for measuring most of the SAQFs, This research has tried to measure them through some measurable Qualitative Criterion. These Qualitative Criteria (we call them QCs) consist of some self-adaptive systems attributes and also some non-adaptive systems qualitative factors. A map between these SAQFs and software systems QCs (which are more measurable) have been introduced. For each QC sufficient metrics for measuring could be dug up based on the problem context. For better knowing about the influence of qualitative factors on each other, a prerequisite and post-requisite graph from relations among SAQFs have been introduced. This relational graph shows the importance and impact of each factors measurement on measuring the systems from self-adaptive viewpoint. For evaluating the method, we have proposed a questionnaire to experts about the model the correctness of these impacts and influences have been verified. In addition, a case study on a system in changing environment evaluated with proposed method and the applicability of the method have been reviewed.
An Evaluation Method for Self-Adaptive Systems / Farahani, Ali; Nazemi, Eslam; Cabri, Giacomo; Rafizadeh, Alireza. - (2016), pp. 2814-2820. (Intervento presentato al convegno 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 tenutosi a Budapest; Hungary nel 9-12 October 2016) [10.1109/SMC.2016.7844665].
An Evaluation Method for Self-Adaptive Systems
Farahani, Ali;CABRI, Giacomo;
2016
Abstract
Nowadays, change is an important aspect of the world. Complexity and change in requirements and environments bring us to Autonomic Systems as a solution. Like any other kind of software system better implementation of a system needs a proper evaluation method for system implementation. To evaluate a Self-Adaptive system, appropriate quality factors are needed for evaluation. This research tries to use non-adaptive systems evaluation methods to evaluate Self-Adaptive ones. The qualitative factors for self-adaptive systems have been extracted from a literature review (as Self-Adaptive System Qualitative Factors or for abbreviation SAQFs). Hence, there is no explicit or even implicit way for measuring most of the SAQFs, This research has tried to measure them through some measurable Qualitative Criterion. These Qualitative Criteria (we call them QCs) consist of some self-adaptive systems attributes and also some non-adaptive systems qualitative factors. A map between these SAQFs and software systems QCs (which are more measurable) have been introduced. For each QC sufficient metrics for measuring could be dug up based on the problem context. For better knowing about the influence of qualitative factors on each other, a prerequisite and post-requisite graph from relations among SAQFs have been introduced. This relational graph shows the importance and impact of each factors measurement on measuring the systems from self-adaptive viewpoint. For evaluating the method, we have proposed a questionnaire to experts about the model the correctness of these impacts and influences have been verified. In addition, a case study on a system in changing environment evaluated with proposed method and the applicability of the method have been reviewed.Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris