Abstract
A stochastic fluid queueing system describes the input-output flow of a fluid in a storage device, called a buffer. The rates at which the fluid enters and leaves the buffer depend on a random environment process. The external governing process is an irreducible CTMC and the fluid from the buffer is emptied at a constant rate μ. Let X(t) denote the buffer content at time t and I(t) denote the state of the random environment at time t. In this paper we present a method for computing the mean first passage times in the (X(t), t ≥ 0) process, as well as in the bivariate ((X(t), I(t)), t ≥ 0) process. We derive a system of first-order non-homogeneous linear differential equations for the mean first passage times which can easily be solved using well-known techniques. The method developed here can be readily implemented for computational purposes. We present two examples illustrating how to find explicitly the analytical solution to a small two-state problem and how to obtain numerical solutions to a multistate problem.
Original language | English (US) |
---|---|
Pages (from-to) | 308-318 |
Number of pages | 11 |
Journal | Operations Research Letters |
Volume | 30 |
Issue number | 5 |
DOIs | |
State | Published - Oct 2002 |
Keywords
- Buffer
- First passage times
- Fluid models
ASJC Scopus subject areas
- Software
- Management Science and Operations Research
- Industrial and Manufacturing Engineering
- Applied Mathematics