The main goal of a problem solving activity is to improve the performance of a system by means of enhancing a useful function or by eliminating a harm. In both situations the first issue to face is to identify the key components to work on and the right level of detail to use in order to effectively model the system and successfully solve the problem. Recent studies have demonstrated that from 3000 raw ideas only 1 become a commercial success. This is why a low accuracy or an insufficient attention during the problem formulation and definition may affects the goodness of the final solution.
Due to the growing interest in innovation issues, support for suitable problem definition and reformulation is increasingly important and needed. An answer to this request is BOB-UP, a procedure implemented by software to drive the user to reformulate an initial problem from natural language into a well defined and structured reformulation, allowing subsequent solutions without getting lost on useless roads (a problem well defined is virtually solved). Such a name was chosen to emphasize the role it has in emerging and surfacing the right problem, regardless of the complexity of the technical system considered.
This work takes into account such a long experience in problem solving methods, especially root cause analysis and TRIZ. Its task is to guide the user to select the critical element of the system. It tends to a clear reformulation of the starting problem in a more suitable way to solve it.
Framework of the procedure is based on an inedited cause-effect analysis, articulated in 3 reformulation steps. For completing each step design ontologies, linguistic rules, modelling tools and sketches are carefully integrated. The output of such a computer aided reformulation, consists of a well defined problem reformulation. Unlike most famous and traditional cause-effect approaches, in BOB-UP environment the user is guided step by step in identifying the best detail level of description, focusing the undesired effect(s), and finally identifying the crucial element on which an intervention is needed. The role of the software is to help fill the reference problem reformulation model explaining how to do it, giving definitions and rules to better understand what is needed, and to provide tools and advices for the most complicate and divergent steps.
The software has been tested on several case study performed by students and by a certain number of industrial applications. Even if its effectiveness is demonstrated the procedure is continually improved. The algorithm is now under patent pending and a further experimental campaign is still in progress.
