The Excel functions discussed in this chapter are useful for solving a lot of probability problems, but there are other problems that, even though they are similar to normal or binomial problems, cannot be solved with these functions. In cases like this, simulation can often be used. Here are a couple of simulate 500 replications of the experiment.
a. You observe a sequence of parts from a manufacturing line. These parts use a component that is supplied by one of two suppliers. Each part made with a component from supplier 1 works properly with probability 0.95, and each part made with a component from supplier 2 works properly with probability 0.98. Assuming that 100 of these parts are made, 60 from supplier 1 and 40 from supplier 2, you want the probability that at least 97 of them work properly.
b. Here we look at a more generic example such as coin flipping. There is a sequence of trials where each trial is a success with probability p and a failure with probability 1 - p. A run is a sequence of consecutive successes or failures. For most of us, intuition says that there should not be long runs. Test this by finding the probability that there is at least one run of length at least six in a sequence of 15 trials. (The run could be of 0s or 1s.) You can use any value of p you like—or try different values of p.