Due to its' connectionist nature, abductive reasoning may get neural network implementations that yet require structure adaptation to the abduction problems which Bylander and the team asserted. The paper proposes neural models for all known abduction problems, in a really unified manner, and with a sound and straightforward embedding in the existing neural network paradigms.

The chapter focuses on the fault diagnosis of artefacts often met in industry (and not only), that executes mainly functions involving conductive flows of matter and energy, i.e. multifunctional conductive flow systems” (MCFSs). The proposed MCFS abstraction is close to human diagnosticians’ way of conceiving entities and relations on physical, functional and behavioural structures. Diagnosis reasoning is intrinsically abductive reasoning. This

A complex system in industry is often a conductive flow system. Its abnormal behaviour is difficult to manage due to incomplete and imprecise knowledge on it, also due to propagated effects that appear at faults. Human experts use knowledge from practice to represent abnormal ranges as interval values but they have poor knowledge on variables with no direct link to target system’s goals. The paper proposes a new fuzzy arithmetic, suited to calculate

Knowledge involved in diagnosis of real complex systems comes from human experts and requires appropriate discrete and qualitative representation. The large amount of information resulted is difficult to be managed and prepared to enter the diagnosis system without the help of an appropriate tool. The paper proposes a knowledge elicitation scheme for multifunctional conductive flow systems under fault diagnosis along with appropriate representation