Qualitative-Quantitative Bayesian Belief Networks for reliability and risk assessment

Abstract

This paper presents an extension of Bayesian belief networks (BBN) enabling use of both qualitative and quantitative likelihood scales in inference. The proposed method is accordingly named QQBBN (Qualitative-Quantitative Bayesian Belief Networks). The inclusion of qualitative scales is especially useful when quantitative data for estimation of probabilities are lacking and experts are reluctant to express their opinions quantitatively. In reliability and risk analysis such situation occurs when for example human and organizational root causes of systems are modeled explicitly. Such causes are often not quantifiable due to limitations in the state of the art and lack of proper quantitative metrics. This paper describes the proposed QQBBN framework and demonstrates its uses through a simple example.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.