Models of random machine downtimes for simulation. (manufacturing systems) (column)

Models Of Random Machine Downtimes For Simulation

The most important source or randomness for many manufacturing systems is that associated with machine breakdowns or unscheduled downtime results from such events as actual machine failures, part jams, and broken tools. The following example illustrates the importance of modeling machine downtime correctly in simulation studies.

A company is going to buy a new machine tool from a vendor who claims that the machine will be down 10 percent of the time. However, the vendor has no data on how long the machine will operate before breaking down or on how long it will take to repair the machine. Some simulation analysts have accounted for random breakdowns by simply reducing the machine processing rate by 10 percent. We will see, however, that this can produce results that are quite inaccurate.

Suppose that the single-machine-tool system will actually operate in accordance with the following assumptions when installed by the purchasing company:

* Jobs arrive with exponential

interarrival times with a mean

of 1.25 minutes.

* Processing times for a job at

the machine are a constant 1

minute.

* The machine operates for an

exponential amount of time …