Metodi e strumenti per la modellazione e soluzione inventiva di problemi complessi

Problem solving methods, such as TRIZ [1], offer an organized paradigm to approach a technical issue an to provide ideas for new solutions. According to the trend of anticipating the detection of errors along the product development process to shorten cycle times, in problem solving practices the correct formulation of the right problem to solve is the crucial activity. Actually, most of the failures occurring to gather good solutions depend on actions done before to the real solving steps, that are then conducted on the basis of wrong information and may bring to nowhere or, in the best case, may solve the wrong problem. The failure is mostly caused by two reasons: an ill modeling of the system or a wrong definition of goals. Both may be partially ascribed to traditional tools supporting product/problem analysis that show some limits when approaching complexity or when in the system process aspects prevails on product ones. The aim of this work is to present a deep analysis of the phase preceding problem solving and to describe the tools, both of classical TRIZ and brand new, supporting those activities. The research has been accomplish along with Product Development centre of Bticino s.p.a. that has provided a case study and technical support. Classical TRIZ tools devoted to process/problem modeling are Functional Analisys and Substance-Field (Su-Field). The first has its roots in Lawrence Miles’ Value Analysis [2] and is very powerful for product decomposition and mapping of useful or harmful functions, but shows some limits for complex systems or process modeling. Su-Fields models are very elementary models representing two elements (tool and object) and the field connecting them. They are almost only used in ARIZ algorithms and are not useful to map the knowledge concerning a whole system. Some TRIZ application are based on event models, derived mainly from cause and effect modeling technique and focused on the root cause or “root conflict”. Valeri Souchkov’s RCA+ [3] is a good example of modeling problems to find the right causes and the connections provoking a failure event. This method give a sharp idea of what happens during the functioning of a device or a process but does not help in describing the system structure and their functions. These drawbacks have been overcome by a newly introduced technique for product/process/problem modeling, called RelEvent Diagram. This diagrams are part of the General Theory of Innovation [4] developed by Greg Yezersky within the Institute for Professional Innovators settled in Detroit. To practically show pros and cons of Functional Analysis, RCA+ and RelEvent Diagrams, they have been all used in case study. Results of the modeling activities will be shown, together with some of the ideas obtained by problem solving activity with classical TRIZ tools.