The Use of Probabilistic Models and Simulation to Support Efficient and Transparent Decision Making

2016 ◽  
Author(s):  
Brian Zacour
Keyword(s):  
Decision Making ◽  
Probabilistic Models

Probabilistic Machine Learning for improved Decision-making with 3-D Geological Models

2021 ◽  
Author(s):  
Florian Wellmann ◽  
Miguel de la Varga ◽  
Nilgün Güdük ◽  
Jan von Harten ◽  
Fabian Stamm ◽  
...  
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Imperfect Information ◽  
Probabilistic Models ◽  
Uncertainty Propagation ◽  
Model Parameters ◽  
Data Types ◽  
Subsurface Structures ◽  
Geological Models ◽  
Probabilistic Machine Learning

<p>Geological models, as 3-D representations of subsurface structures and property distributions, are used in many economic, scientific, and societal decision processes. These models are built on prior assumptions and imperfect information, and they often result from an integration of geological and geophysical data types with varying quality. These aspects result in uncertainties about the predicted subsurface structures and property distributions, which will affect the subsequent decision process.</p><p>We discuss approaches to evaluate uncertainties in geological models and to integrate geological and geophysical information in combined workflows. A first step is the consideration of uncertainties in prior model parameters on the basis of uncertainty propagation (forward uncertainty quantification). When applied to structural geological models with discrete classes, these methods result in a class probability for each point in space, often represented in tessellated grid cells. These results can then be visualized or forwarded to process simulations. Another option is to add risk functions for subsequent decision analyses. In recent work, these geological uncertainty fields have also been used as an input to subsequent geophysical inversions.</p><p>A logical extension to these existing approaches is the integration of geological forward operators into inverse frameworks, to enable a full flow of inference for a wider range of relevant parameters. We investigate here specifically the use of probabilistic machine learning tools in combination with geological and geophysical modeling. Challenges exist due to the hierarchical nature of the probabilistic models, but modern sampling strategies allow for efficient sampling in these complex settings. We showcase the application with examples combining geological modeling and geophysical potential field measurements in an integrated model for improved decision making.</p>

Probability of Exceedance (POE) Methodology for Developing Integrity Programs Based on Pipeline Operator-Specific Technical and Economic Factors

2002 ◽  
Author(s):  
Rafael G. Mora ◽  
Curtis Parker ◽  
Patrick H. Vieth ◽  
Burke Delanty
Keyword(s):  
Decision Making ◽  
Probabilistic Models ◽  
Cost Benefit Analysis ◽  
Management Program ◽  
Metal Loss ◽  
Decision Making Process ◽  
Pipeline System ◽  
Probability Of Exceedance ◽  
Integrity Verification ◽  
Inspection Data

With the availability of in-line inspection data, pipeline operators have additional information to develop the technical and economic justification for integrity verification programs (i.e. Fitness-for-Purpose) across an entire pipeline system. The Probability of Exceedance (POE) methodology described herein provides a defensible decision making process for addressing immediate corrosion threats identified through metal loss in-line inspection (ILI) and the use of sub-critical in-line inspection data to develop a long term integrity management program. In addition, this paper describes the process used to develop a Corrosion In-line Inspection POE-based Assessment for one of the systems operated by TransGas Limited (Saskatchewan, Canada). In 2001, TransGas Limited and CC Technologies undertook an integrity verification program of the Loomis to Herbert gas pipeline system to develop an appropriate scope and schedule maintenance activities along this pipeline system. This methodology customizes Probability of Exceedance (POE) results with a deterministic corrosion growth model to determine pipeline specific excavation/repair and re-inspection interval alternatives. Consequently, feature repairs can be scheduled based on severity, operational and financial conditions while maintaining safety as first priority. The merging of deterministic and probabilistic models identified the Loomis to Herbert pipeline system’s worst predicted metal loss depth and the lowest safety factor per each repair/reinspection interval alternative, which when combined with the cost/benefit analysis provided a simplified and safe decision-making process.

scholarly journals Analysis of Nonlinear Relationships in Dual Epidemics, and Its Application to the Management of Grapevine Downy and Powdery Mildews

2009 ◽  
Vol 99 (8) ◽  
pp. 930-942 ◽  
Author(s):  
Serge Savary ◽  
Lionel Delbac ◽  
Amélie Rochas ◽  
Guillaume Taisant ◽  
Laetitia Willocquet
Keyword(s):  
Decision Making ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Powdery Mildew ◽  
Disease Control ◽  
Probabilistic Models ◽  
Nonlinear Modeling ◽  
Powdery Mildews ◽  
Cropping Season ◽  
Tactical Decision

Dual epidemics are defined as epidemics developing on two or several plant organs in the course of a cropping season. Agricultural pathosystems where such epidemics develop are often very important, because the harvestable part is one of the organs affected. These epidemics also are often difficult to manage, because the linkage between epidemiological components occurring on different organs is poorly understood, and because prediction of the risk toward the harvestable organs is difficult. In the case of downy mildew (DM) and powdery mildew (PM) of grapevine, nonlinear modeling and logistic regression indicated nonlinearity in the foliage–cluster relationships. Nonlinear modeling enabled the parameterization of a transmission coefficient that numerically links the two components, leaves and clusters, in DM and PM epidemics. Logistic regression analysis yielded a series of probabilistic models that enabled predicting preset levels of cluster infection risks based on DM and PM severities on the foliage at successive crop stages. The usefulness of this framework for tactical decision-making for disease control is discussed.

scholarly journals MODELING GOAL-ORIENTED DECISION MAKING THROUGH COGNITIVE PHASE TRANSITIONS

2009 ◽  
Vol 05 (01) ◽  
pp. 143-157 ◽  
Author(s):  
ROBERT KOZMA ◽  
MARKO PULJIC ◽  
LEONID PERLOVSKY
Keyword(s):  
Decision Making ◽  
Phase Transitions ◽  
Cognitive Processing ◽  
Probabilistic Models ◽  
Temporal Dynamics ◽  
Mathematical Tool ◽  
Neural Populations ◽  
Spatio Temporal ◽  
Experimental Findings ◽  
High Level

Cognitive experiments indicate the presence of discontinuities in brain dynamics during high-level cognitive processing. Non-linear dynamic theory of brains pioneered by Freeman explains the experimental findings through the theory of metastability and edge-of-criticality in cognitive systems, which are key properties associated with robust operation and fast and reliable decision making. Recently, neuropercolation has been proposed to model such critical behavior. Neuropercolation is a family of probabilistic models based on the mathematical theory of bootstrap percolations on lattices and random graphs and motivated by structural and dynamical properties of neural populations in the cortex. Neuropercolation exhibits phase transitions and it provides a novel mathematical tool for studying spatio-temporal dynamics of multi-stable systems. The present work reviews the theory of cognitive phase transitions based on neuropercolation models and outlines the implications to decision making in brains and in artificial designs.

Models of Quality Estimation of Multi-Parametric Management of Complicated Systems

2018 ◽  
pp. 161-172
Author(s):  
O. V. Morozova
Keyword(s):  
Decision Making ◽  
Probabilistic Models ◽  
Business Processes ◽  
Quantitative Estimation ◽  
Integral Functional ◽  
Computer Experiment ◽  
Mathematic Model ◽  
Practical Applications ◽  
Parametric System ◽  
The Impact

The author puts forward methodology of estimating and forecasting control quality in the system of managing business processes of the complicated multi-parametric system. The results of estimation are given in qualitative and quantitative forms. In order to calculate errors in control probabilistic models of false and unfound rejects were developed. For qualitative estimation of the system functioning imprecise models were developed. Probabilistic models make it possible to study the impact of statistic characteristics of modeling agents on control errors and risks. Producer’s risk and customer’s risk are considered as risks. Truthfulness and effectiveness of modeling can be checked through computer experiment on the basis of imitation algorithm. The mathematic model and the imitation algorithm have universal nature and can be used in different scientific and technical practical applications. The article describes a concrete example of estimating risks of decision-making in personnel quality management in higher education. To do this the theory of imprecise multitudes is used. To estimate the personnel quality a differentiated approach by totality of such parameters as experience, education, qualification, health, age is applied. As these parameters can hardly be estimated quantitatively, the imprecise approach on the basis of linguistic indicators is used. To combine differentiated indicators in the final integral assessment ‘human resource’ the mathematic expression was put forward. The author advanced a new multi-approach methodology of quantitative estimation of risks of decision-making in multi-parametric system of control and management through differentiated and integral functional indicators of the unit.

scholarly journals Probabilistic Rationale of Actions for Artificial Intelligence Systems Operating in Uncertainty Conditions

2019 ◽  
Vol 1 (2) ◽  
Author(s):  
Andrey Ivanovich Kostogryzov
Keyword(s):  
Artificial Intelligence ◽  
Decision Making ◽  
Probability Distribution ◽  
Probabilistic Models ◽  
System Analysis ◽  
Optimization Methods ◽  
Distribution Functions ◽  
Probability Distribution Functions ◽  
Prognostic Modeling ◽  
Artificial Intelligence Systems

The approach for probabilistic rationale of artificial intelligence systems actions is proposed. It is based on an implementation of the proposed interconnected ideas 1-7 about system analysis and optimization focused on prognostic modeling. The ideas may be applied also by using another probabilistic models which supported by software tools and can predict successfulness or risks on a level of probability distribution functions.  The approach includes description of the proposed probabilistic models, optimization methods for rationale actions and incremental algorithms for solving the problems of  supporting decision-making on the base of monitored  data and rationale a robot actions in uncertainty conditions. The approach means practically a proactive commitment to excellence in uncertainty conditions. A suitability of the proposed models and methods is demonstrated by examples which cover wide applications of artificial intelligence systems.

Quantitative methods in the field of economic sciences

VUZF Review ◽  
2020 ◽  
Vol 5 (2) ◽  
pp. 22-37
Author(s):  
Maria Borowska
Keyword(s):  
Decision Making ◽  
Mathematical Models ◽  
Quantitative Methods ◽  
Service Management ◽  
Probabilistic Models ◽  
Science Research ◽  
Complex Nature ◽  
Economic Science ◽  
Mathematical Methods ◽  
Product Delivery

The usage of quantitative tools for creating the mathematical models of functioning different economic facilities abounds the opportunity for better understanding and acquaintance of the surrounding reality. A lot of thinkers identify even universality of the particular branch of knowledge with the extent of its ‘mathematization’. Applying mathematical methods so called quantitative provide great and not to overestimate services not only in the science research of technique, physics, astronomy, biology and medicine, but also – within the qualitative methods- in the field of social science in the sphere of the control of the quality of production or in the process of service management or decision making. Complex nature of the social and economic phenomena requires making the usage of the most modern means and the ubiquitous computerization significantly confirms the usefulness of these methods. Progressing ‘mathematization’ and computerization of the science forces creating and applying quantitative (mathematical) models including economic science. The model of operating of studied system was considered in two variants. I. when the process of the product delivery to the store represents inclusively the subsystem of production and the subsystem of the transportation – it could be then said that the level of filling the store up is controlled by the aggregated process of the delivery of the product.  when the process of the product delivery to the store takes into account explicate both the production process and also the operating of transportation subsystem, so it is then the structural process of the product delivery. Both in the aggregated and structural version, the analyses of the functioning of the system was made in three variants of the store filling: intermediate state of the store filling; zero state of the store filling that is lower barrier; the state of full storage of the store, that is the upper barrier. The result of my analyses are two proprietary probabilistic models of system operation which are presented through the system of differential equations both in the aggregated and structural variant. Probabilistic models of functioning of the system in both variants presented throughout the probabilistic model also enable determining sizing prognosis which are characteristic for the functioning of this system. These prognoses are transferred to the unit of the management system and they provide the premises to the streamline of its functioning. These tools create the basics of theoretical and methodological constructed computer programmes of the informative systems of decision-making support.

scholarly journals Embedded Bayesian Network Contribution for a Safe Mission Planning of Autonomous Vehicles

Algorithms ◽  
2020 ◽  
Vol 13 (7) ◽  
pp. 155 ◽  
Author(s):  
Catherine Dezan ◽  
Sara Zermani ◽  
Chabha Hireche
Keyword(s):  
Decision Making ◽  
Autonomous Vehicles ◽  
Probabilistic Models ◽  
Automatic Generation ◽  
Mission Planning ◽  
Pattern Design ◽  
Decision Network ◽  
Markov Decision ◽  
Speed Up ◽  
Decision Making Model

Bayesian Networks (BN) are probabilistic models that are commonly used for the diagnosis in numerous domains (medicine, finance, transport, robotics, …). In the case of autonomous vehicles, they can contribute to elaborate intelligent monitors that can take the environmental context into account. We show in this paper some main abilities of BN that can help in the elaboration of fault detection isolation and recovery (FDIR) modules. One of the main difficulty with the BN model is generally to elaborate these ones according to the case of study. Then, we propose some automatic generation techniques from failure mode and effects analysis (FMEA)-like tables using the pattern design approach. Once defined, these modules have to operate online for autonomous vehicles. In a second part, we propose a design methodology to embed the real-time and non-intrusive implementations of the BN modules using FPGA-SoC support. We show that the FPGA implementation can offer an interesting speed-up with very limited energy cost. Lastly, we show how these BN modules can be incorporated into the decision-making model for the mission planning of unmanned aerial vehicles (UAVs). We illustrate the integration by means of two models: the Decision Network model that is a straightforward extension of the BN model, and the BFM model that is an extension of the Markov Decision Process (MDP) decision-making model incorporating a BN. We illustrate the different proposals with realistic examples and show that the hybrid implementation on FPGA-SoC can offer some benefits.

Sign in / Sign up

Export Citation Format

Share Document