Uncertainty Methods in Expert Systems

Abstract

The paper introduces the main measures of uncertainty and discusses some of the problems associated with conventional uncertainty propagation methods, such as those based on independent probabilities and the methods used with certainty factors, belief, and possibility functions. Some examples show the importance of the associated errors. Alternative methods, such as log‐linear regression and causal networks or influence diagram models, are discussed. Finally, their structural and parametric learning possibilities are analyzed.

Application of the Modified HAZOP to an Advisory System Using Rule–Based Approach for Packed Column

“…In general, expert system can be defined as a computer program that encodes the knowledge of an expert in some specific areas and uses inference procedures in solving problems that are difficult enough to require significant human expertise for their solution. Specifically, an expert system is a computer system that simulates the learning, memorization, reasoning, communication, and action processes of a human expert in a given area of science, giving in this way a consultant that can substitute a human expert with reasonable guaranties of success [5]. Therefore, the ultimate aim of every expert system is the substitution of the human expert and improvement on their performance.…”

“…Domain-specific knowledge comprises of the expert's factual and heuristic, rule-oriented, and structure-oriented knowledge. Meanwhile, the other represents general problem-solving strategies applicable to any set of domain-specific knowledge built into the system [5].…”

“…The scheduler then orders the firing of all such rules. Many schedulers employ meta-rules to provide the principles that determine in what sequence rules should be fired in [5].…”

“…The inference engine mechanism employed in the expert system must be compatible with the data structure form used to represent the domain knowledge. One of the key features of expert system technology is the distinct separation of the domain-specific and domain-independent aspects of knowledge, allowing much greater ease in updating or modifying the knowledge base [5].…”

Design with Uncertainty: The Role of Future Options for Infrastructure Integration

“…For events that depend on uncertain elements, predictions cannot be made in a deterministic manner. In these cases, approaches that take uncertainty in consideration, such as probabilistic approaches (Castillo and Alvarez, ), have to be used.…”