Devising Configuration Guidance with Process Mining Support
Résumé
[Context and Motivation] Adopting a Product Line strategy can be a double-edged sword when the configuration process proves complex. In fact, complex configuration processes deprive industry and stakeholders of the benefits of Product Line Engineering; mainly, decreased time to market and increased customer satisfaction. [Question/Problem] Unfortunately, this stumbling block is widespread mainly when stakeholders deal with large product line models that contain a plethora of variants. Therefore, the configuration process becomes a monotonous and an error-prone task. Consequently, it is of great interest to industry that product line configuration modules are able to recommend the best configuration alternatives until leading the stakeholders to a satisfying configuration experience. [Principal idea/Contribution] From this perspective, this research aims to enhance the new generation of product line configuration modules, maximizing the stakeholders comfort based on previous configuration processes. Therefore, previous configuration processes are logged, mined and interpreted to produce guidance solutions. Afterwards, guidance solutions are devised and orchestrated by the CONFILOG method detailed in this paper. [Results] Experiments of CONFILOG show that configuration guidance can be produced with process mining support. Solutions feasibility and distinction is demonstrated through an Empirical Study With Students applied to a bike product line.
Origine | Fichiers produits par l'(les) auteur(s) |
---|