Useful Statistical Models

1. Queueing system. This is one of the most often used models. We will study queueing system in great detail later. Queueing system can have a lot of variations that can be used to model many real life situation. We have seen a few examples in earlier chapters.

   Queueing systems often include random variables such as service time at various server(s); inter-arrival time for various customer streams; routing distribution of customers.

   Example 6.8, 6.9 on page 193. Emphasize on the case where empirical data can produce histogram which shows the trend, often close to a mathematical model.

2. Inventory system. This is another model used to simulation storage problems, inventory problems. A typical inventory has three random variables: the number of units demanded per order or per time period; the time between demands; the lead time.

3. Reliability and maintainability. Time to failure of a component, of a system.

4. Limited data. When complete data is not avaiable, three distributions can be used to model the system, uniform, triangular, and beta distribution.

5. Other distributions. A few other distributions can be used to model systems that are not covered above: Bernoulli and binomial distributions are for discrete system; hyperexponential may be useful in other instances.