Example Models
AgenaRisk comes with a large number of example models that illustrate the different kinds of problem that can be tackled.
The models are divided into the following categories: Introductory (i.e. used or referred to in the AgenaRisk Tutorials), Basic, Advanced. To view and run these models you will need to install the evaluation version of AgenaRisk. But we also have examples of models in action.
Introductory Models
| Name | Description |
| Asia | Diagnoses tuberculosis, lung cancer and bronchitis based on various factors including whether the patient in question has been to Asia recently |
| Car Costs | Predicts the long-term running costs of owning different makes of car |
| Flood Model | A temporal model that illustrates how risk can be represented causally |
| Hypothesis Testing | Statistically tests the hypothesis that one material is more faulty than another |
Basic Models
| Name | Description |
| Aggregating Distributions | Calculates the total cost of a product based upon the uncertain costs of its components |
| Causal Risk Register | Determines whether a project is at risk by reasoning about the causes and consequences of risk |
| Constraint Satisfaction | Solves the classic "map colouring" problem, where a map has to be filled in with four colours in such a way that no adjacent region can contain the same clour |
| Dependent Coin Flips | Two simple probability experiments that involve flipping coins |
| Fault Tree Analysis | Two examples of how AgenaRisk can be used to construct a fault tree for analysing the reliability of complex systems |
| Fire | Models a simple fire alarm system and shows the principle of "explaining away" evidence |
| Intensive Care Monitoring | A model for monitoring patients in intensive care |
| KUUUB | Adjusts quantitative loss predictions using qualitative KUUUB (Known Unknown, Unknown Unknown and Bias) data |
| Monty Hall Dilemma | An illustration of the classic game show dilemma in which a contestant tries to find a prize by opening doors |
| Mountain Pass | Decision analysis example in which a man has to decide how he should travel to an appointment |
| Naive Bayesian Classification | Uses existing data about known banks to characterise and thus predict the loss distributions of unknown banks |
| Noisy Or | A simple model that shows the effects of liver failure and hepatitis on jaundice and that demonstrates the Noisy Or function |
| Parameterised Distribution using Constants | Illustrates how constants can be used to change the parameters of distributions without needing to edit the corresponding expression or regenerate the corresponding NPT |
| Printer Fault Diagnosis | A model for predicting and diagnosing printer faults |
| Risk Control Self Assessment | Predicts losses in a business based upon the presence and absence of different risk controls |
| Risk Drivers and Indicators | Two models that show the correlation between risk drivers and risk indicators |
| River Flooding | A time-series model showing how the flooding of a river is influenced by rainfall and the state of its flood defences |
| Safety | A model used for assessing the safety of a critical system |
| Simple Testing Process | Illustrates how the quality of the testing process can influence the number of defects found during testing |
| Statistical Distributions | Illustrates the range of different statistical distributions supported by AgenaRisk |
| Wet Grass | A simple example that shows how a single event can occur as a result of two different causes |
Advanced Models
| Name | Description |
| Batch Learning Model | A version of the Diet Experiment model that applies data in batches to improve computational efficiency |
| Biased Coin Flip Experiment | Shows how coin flip data can be used to assess the bias of a biased coin |
| Diabetes Treatment | A model for advising practitioners about adjusting insuling doses |
| Diet Experiment | Determines whether there are statistically significant differences between the effects that four different diet plans have on the level of blood coagulation in patients |
| Hailfinder | A model for forecasting severe summer hail in an area of Colorado |
| Information Fusion | Predicts the type of an enemy unit based upon information received through various sensors |
| Mildew | A model for deciding the amount of fungicide to be used against mildew in wheat |
| Mixing Product Failure Data | Uses failure data from a set of products to predict failure in other similar products |
| Operational Risk in Finance | Predicts operational risk using aggregated loss data and causal information about loss events, indicators and controls |
| Reliability Estimation | Predicts the failure rate of a system based upon failures observed, test duration and the subjective judgments of engineers |
| Six Sigma Defect Containment | A model that predicts the defects introduced into a piece of software and the likely probability of detecting and containing these defects based on an evaluation of the design and test process quality |
| Six Sigma Testing and Rework | Predicts the testing effectiveness, rework effectiveness and number of residual defects involved in a software process |
| Software Defect Prediction | Illustrates and explains a number of classical empirical studies comparing 'pre' and 'post’ software release defects. |
| Software Project Risk | A complex model that captures the factors determining the success of a software project |
| Water Purification | A time-series model representing the biological processes of a water purification plant |
Models in action
| Name | Description |
| Investment Risk | Demonstrates some of the power of Bayesian nets and the AgenaRisk tool using a simple example of stock market investment. [Click here toview. (Windows Media Video)] |
| Software Defects Prediction | Illustrates and explains a number of classical empirical studies comparing 'pre' and 'post’ software release defects. [Click here to view.(Powerpoint Slide)] |
| Software Project Risk | This demonstrates a phased software project risk model in action. The focus is on defect prediction. [Click here to view. (Flash Video will require you to accept the security warning.)] |
| Software Project Trade-off Model | This model can be used by managers early in a software project for resource prediction, trade-off analysis and risk assessment. The video shows a typical scenario and how you can make predictions with very little information and then revise them given new project constraints. [Click here to view. (Windows Media Video)] |