Purpose The purpose of this paper is to raise awareness among manufacturing researchers and practitioners the potential Bayesian networks (BNs) enhance decision making in those parts domain where uncertainty a key characteristic. In doing so, describes development an intelligent support system (DSS) help operators Motorola diagnose correct faults during process product testing. Design/methodology/approach combines BNs user interface produce multi‐media advice for operators. Findings Surveys...

In this paper, we consider the application of dynamic Bayesian networks to prognostic modelling equipment in order better inform maintenance decision-making. We provide a brief overview and their reliability modelling. An example is then provided which an considered be one six states there are two imperfect condition monitoring indicators available evidence about equipment's true state tends deteriorate over time. With example, show how decays time situation where repair not possible simple...

AbstractThe potential of hybrid models to enhance simulations the real world is explored. While scope for design such large, focus here brings together agent-based (AB) and system dynamics (SD) modelling within a defined architectural framework. Comprising number modules, each which implemented in single paradigm, looks exploit from range approaches tools. Coded programming environment, international diffusion technological innovation used as case study highlight simulation model...

Simulation metamodels have been used for optimization, prediction, sensitivity analysis and understanding of complex, real-world systems. Since most simulation models contain a large number input parameters, it is great interest to determine the important ones include in metamodel given particular modeling context, i.e. set questions which are be addressed by metamodel. This paper employs Morris' randomized one-at-a-time (OAT) design as factor screening method prior developing metamodels....

The analytic standards governing the production of intelligence are outlined in a number Intelligence Community Directives (ICDs). In this paper, we concerned with ICDs 203, 206 and 208 and, particular, how these relate to handling uncertainty estimative intelligence. An inductive thematic analysis is employed which identifies several recurring themes. addition, conceptual map developed highlights relationships level inter-connectedness between standards. Requirements for improved...

In this paper, we discuss the use of multi-layer perceptron (MLP) neural networks to predict outputs from a stochastic combat simulation model. The faster, auxiliary model approximate input-output relationships complex is known in community as metamodelling. Since many models contain large number input parameters, however it necessary reduce set considered and determine most important ones include metamodel given particular modelling context. This paper employs two-stage experimental design...

In operational engineering maintenance situations, limitations on time, resource or the information available often inhibit rigorous analysis complex decision problems. Decision-makers who are compelled to act in such circumstances, may be informed by some level of if available, else have rely their unsupported judgement. This paper presents three risk decision-making case studies across a 20 year span from rail, aerospace, and military aviation contexts, highlighting fallibilities using...

Assessing the credibility of an evidential source is important part intelligence analysis, particularly where human concerned. Furthermore, it frequently necessary to combine multiple items evidence with varying credibilities. Bayesian networks provide a powerful probabilistic approach fusion information and are increasingly being applied in wide variety settings. In this paper we explore their application analysis simple example concerning potential attack on infrastructure target. Our main...

Soft tissue simulants are traditionally used to provide a post impact medium suitable for replicating human anatomy. The information provided within this study provides both economic and environmental assessment compares its results against material performance.

Simulation metamodels have been used for optimization, prediction, sensitivity analysis and understanding of complex, real-world systems. Since most simulation models contain a large number input parameters, it is great interest to determine the important ones include in metamodel given particular modeling context, i.e. set questions which are be addressed by metamodel. This paper employs Morris' randomized one-at-a-time (OAT) design as factor screening method prior developing metamodels....

Important assessments of events and activities relating to military, terrorist hybrid adversaries the intentions foreign governments, are made every day, usually involving subjective or 'estimative' probabilities an associated level confidence. The way in which these uncertainties accessed communicated can potentially have enormous impact consequences. Challenges reinforced by increasingly complex intelligence problems for contemporary analytic paradigm is not tailored cope. It important...

Abstract In this paper, Markovian models of three‐on‐one stochastic firefights between ground‐based weapon systems are developed. These address a common scenario interest to the military, but one which has been much neglected in analytic combat models, that hidden defender and an exposed attacking force. Each combatant must detect opponent before commencing their firing cycle, task is considerably more difficult for attacker. developed here, detects group after exponentially distributed time...

Abstract Countering improvised explosive devices (C-IED) is a significant theme of the twenty-first century, particularly in regions with limited governance and fragile rule law. Many strands activity are involved, human interaction proving difficult to predict. However, Bayesian belief updating (used across several academic fields provide insight into behaviours) has never been considered. Given breadth C-IED, this research focusses on state affected by conflict, where illicit diversion...

The range of books reviewed is wide, covering theory and applications in operations research, statistics, management science, econometrics, mathematics, computers, information systems (no software reviewed). In addition, we include other fields that emphasize technical applications. Publishers who wish to have their should send them Professor Benjamin Lev. We list the received; not all received can be because space time are limited. Those would like review urged me names, addresses, specific...