# In the rst part of this section we will consider

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ng-run cost per unit time is Eri 10 + 0.3T = Eti T 0.05T 2 To maximize we take the derivative d Eri 0.3(T 0.05T 2 ) (10 + 0.3T )(1 0.1T ) = dT Eti (T 0.1T 2 )2 0.3T 0.015T 2 10 0.3T + T + 0.03T 2 = (T 0.1T 2 )2 The numerator is 0.015T 2 + T 10 which is 0 when p p 1 ± 1 + 4(0.015)(10) 1 ± 1.6 T= = 2(0.015) 0.03 We want the + root which is T = 8.83. Using the idea of renewal reward processes, we can easily treat the following extension of renewal processes. Example 2.5. Alternating renewal processes. Let s1 , s2 , . . . be independent with a distribution F that has mean µF , and let u1 , u2 , . . . be independent with distribution G that has mean µG . For a concrete example consider the machine in Example 1.1 that works for an amount of time si before needing a repair that takes ui units of time. However, to talk about things in general we will say that the alternating renewal process spends an amount of time si in state 1, an amount of time ui in state 2, and then repeats the cycle again. Theorem 3.4. In an alternating renewal process, the limitin...
View Full Document

## This document was uploaded on 03/06/2014 for the course MATH 4740 at Cornell University (Engineering School).

Ask a homework question - tutors are online