Estimation des non-fonctionnels: nous avons besoin de votre aide

Estimating non-functional requirements has been an issue for decades. Nous avons essayé le 14 GSC pour corriger la taille fonctionnelle des complications non fonctionnelles, Mais ça n'a pas marché. Nous avons essayé des approches de type COCOMO, which gave unsatisfactory results. We have tried SNAP, but that is far too complicated. So the best approach at this moment is to ignore them and let the contractmanagers fix the problem or point at the architects to come up with an estimate.

A small group of measurement specialists and architects has now come up with a framework that might solve the problem. By using principles from Architecture Driven Estimation the architecture of an IT project can be divided into homogeneous parts (functional and non-functional) that can be estimated part by part. This framework needs to be detailed further to become directly usable by IT-professionals who are responsible for estimating IT-projects. We are looking for volunteers who can help us mature this framework. We are already assisted by measurement specialists from ASSEMI from France, but we are still looking for:

  • Architectes de solutions pour nous aider à normaliser les éléments de la structure de répartition de l'architecture dont nous avons besoin
  • Spécialistes des achats pour nous aider à définir des composants standardisés pour exprimer les demandes de propositions de manière standardisée
  • Spécialistes financiers / mesures pour nous aider à définir des méthodes standardisées d'estimation du coût des composants standardisés

Our framework was presented at the recent International Workshop on Software Measurement in Gothenburg (Suède). You can find it ici.

The IWSM presentation is available ici.

When you are interested to help us in any way, please send an e-mail to

Partager cet article sur:

Laisser une réponse