Figure 1 - uploaded by Myriam Merad
Content may be subject to copyright.
Source publication
In the field of risk management, decision-aiding is often confused with and reduced to a problem of assessment. The decision-aiding cycle (canonical model) presented in this paper consists in several steps ranging from 'stating and framing the decision problem' to 'working out recommendations'. The kernel of the canonical model relies on two main s...
Contexts in source publication
Context 1
... step A has been performed, the risks related to each alternative have to be assessed in terms of the set of attributes or criteria. This is the aim of this step (Fig. 1b). The very issue of assessment tech- niques raises per se various specific problems exten- sively documented elsewhere. Despite its importance, it goes beyond the scope of this paper. In this case, the crucial point raised concentrates on the nature of information at hand and its consistent processing in the subsequent decision-making ...
Context 2
... or actions on the basis of these vectors would be an intractable, cognitive task for the DM. Indeed, a procedure is needed to aggregate assessments of risks for all the alter- natives so as to compare them. Mathematically speaking, this is the very core of the decision mak- ing process: this step consists in combining subjec- tive information (Fig. 1b) related to the actors and assessments obtained for each alternative (Fig. 1a), with respect to to a given decision making rule or procedure (Fig. ...
Context 3
... task for the DM. Indeed, a procedure is needed to aggregate assessments of risks for all the alter- natives so as to compare them. Mathematically speaking, this is the very core of the decision mak- ing process: this step consists in combining subjec- tive information (Fig. 1b) related to the actors and assessments obtained for each alternative (Fig. 1a), with respect to to a given decision making rule or procedure (Fig. ...
Context 4
... of risks for all the alter- natives so as to compare them. Mathematically speaking, this is the very core of the decision mak- ing process: this step consists in combining subjec- tive information (Fig. 1b) related to the actors and assessments obtained for each alternative (Fig. 1a), with respect to to a given decision making rule or procedure (Fig. ...
Context 5
... from actors, experts and the DM are needed such as tradeoffs or risk appraisals. Although it is sometimes advocated for sake of a mythical objec- tivity, one shall not circumvent these subjective judgments. Therefore, this step is often discarded by practionners. On the contrary, it shall factor in subjective judgments but in an objective manner (Fig. 1d) performed differently with respect to the model of aggregation implemented (see step C). Cord.et al. ...
Context 6
... a mere prescriptive point of view, the last step ( Fig. 1) consists in working out a recommen- dation. The recommendation may have different forms according to the type of problem envisaged p α , p β or p δ . The output information can be an overall score that expresses ranks of alternatives or assignments of objects to predefined classes or indicators that yields just extra information to the ...
Similar publications
Many fine-grained (if not bigger picture) decisions concerning climate change involve significant, even severe, uncertainty. Here, we focus on modelling the decisions of single agents, whether individual persons or groups perceived as corporate entities. We offer a taxonomy of the sources and kinds of uncertainty that arise in framing these decisio...
Citations
... The literature contains different methods to estimate the reliability parameters of a population of equipment from expert knowledge. Where the failure data is insufficient, direct or indirect elicitation processes can be set up with a panel of experts [5,6]. If data is available, Bayesian methods allow prior distribution to be estimated based on expert experience feedback. ...
To meet new needs and respond to changes in the energy market, Hydro-Québec TransÉnergie required new predictive modelling methods and systems to support its asset management activities. It created PRIAD, a robust integration and decision support system. One goal of PRIAD is to assess asset behavior for the purposes of simulating system reliability. A method using a black-box optimization technique was developed to calibrate an expert reliability model with historical data analysis. The model's electrical equipment reliability predictions were satisfactory, but further improvements are planned. One benefit of this approach is to allow experts to reassess their maintenance strategies using modelling results.
