Resource Library

It is important for an insurance company to estimate the amount of claims it will incur in a given year. Here you will find a set of examples to model this problem:
1. A deterministic model to get started.
2. A na•ve @RISK model that fails to capture all the uncertainty in total claims.
3. A better version using the RiskCompound function to capture all of the uncertainty in total claims.
4. A version using the method of resampling to simulate the numbers of claims each year.
Click here to see a video of this example.

An insurance model that illustrates @RISK's Stress Analysis feature.
Click here to see a video of this example.

TopRank recognizes @RISK distribution functions and incorporates them in What-If analyses. This ability provides more flexibility and accuracy in modeling the possible input values in your What-If analysis. In this example, Jupiter Corporation is building a new model of 4-door sedan. Assuming that the car will generate sales for the next 5 years, management has identified 5 factors that can influence the total revenue during that period. Several of these factors have probability distributions associated with them. During a What-If analysis, TopRank sees the probability distributions associated with these items and performs a smart sensitivity analysis using them, stepping through the range of the distribution while spacing the steps such that each interval encompasses equal amounts of probability.

This model represents the process flow of a company's internal sales quotation process. The process is taken from an actual company and has over 36 individual steps involving 10 individuals or departments. For example, when management saw from the simulation results that it took over 24 hours to complete 35 minutes of value-added work, they saw the need for process improvement.

This model is of a manufacturer that needs to reduce the number of defective units produced. It uses @RISK to pinpoint the manufacturing stage that is the worst offender. It also obtains key process capability metrics for each stage, as well as for the entire process, that will help to improve quality and reduce waste. Given historical data, it also uses @RISK's distribution fitting feature to define distribution functions describing the number of defective parts at each stage of the process.

This model is used to calculate the percentage of defective products. Each product component is nondefective if some property of its finished state lies within defined tolerance limits. The project component cells are designated as @RISK outputs with RiskSixSigma property functions defining LSL, USL, and Target value. In this way, you can see graphs of the components' quality and calculate Six Sigma statistics on each component.

This is an extension of the DMAIC (Define Measure, Analyze, Improve, Control) failure model for use in quality control and planning. It includes the use of RiskTheo functions (in this case RiskTheoXtoP) for determining the failure rate without actually running a simulation. The model also illustrates @RISK outputs with RiskSixSigma property functions defining LSL, USL, and Target values for each component.

This model explores an incentive to increase customer loyalty in a market such as the cell phone market. Each year, each of our customers remains with us with a given probability and each of the competitors' customers switches to us with another given probability. The question is whether it is worth our while to incentivize their customers to switch to us. The model assumes a one-time monetary incentive when they make such a switch.

This model is for a company that mails its catalog to a customer every quarter. For each customer, the company keeps track of the recency (the number of catalogs since the customer last purchased) and frequency (the total number of purchases so far). It costs $1 to mail a catalog. If the customer makes a purchase, the company's profit (not counting the cost of mailing the catalog) is Pert distributed with given parameters. The company keeps mailing catalogs to a customer until 24 catalogs produce no purchases, that is, until recency reaches 24.

This model illustrates one possible way oil prices might change through time, as influenced by the market. @RISK's distribution fitting tool is used to simulate future absolute price changes based on historical daily oil prices.

This model is of an oil operator who faces random oil prices and uncertainty in oil volumes over the next five years. To forecast future oil prices, @RISK's Time Series Fit tool is used to fit actual historical oil prices. Then to evaluate a hedge against decreases in oil prices, a "base" model is compared to a model with oil swaps.

This model simulates risks in a network of oil pipelines. There are nine types of risks, and there are nine routes in the pipeline network. Each route has three characteristics: diameter, mean pressure, and distance. For each type of risk and each route, two quantities are simulated: the number of events where that risk type occurs and the typical magnitude of such a risk. These are then accumulated to find the severities of the risk types, by route and total over all routes.