This paper discusses the following model: (i) Arriving customers are accepted in the system in a time interval [t0, t′] and are rejected with compensation after time t′; (ii) the server runs at the cost rate r0 in a time interval [t0, T], t0 ≦ t′ ≦ T, and runs at the increased cost rate r1(r1 > r0) after the closing time T, until the system becomes empty after time t′. The time t′ is called the rejection time. The paper finds the optimal rejection time, that is, the rejection time that minimizes the expected total cost. It can be shown that in certain cases the optimal rejection time can be determined by a nonlinear equation. Numerical examples are given for an M/M/1 queuing system, and the more generalized model is briefly discussed.
[1]
Daniel P. Heyman,et al.
Optimal Operating Policies for M/G/1 Queuing Systems
,
1968,
Oper. Res..
[2]
L. Takács.
Investigation of waiting time problems by reduction to Markov processes
,
1955
.
[3]
N. U. Prabhu,et al.
Application of Storage Theory to Queues with Poisson Arrivals
,
1960
.
[4]
V. E. Benes,et al.
On Queues with Poisson Arrivals
,
1957
.
[5]
D. V. Widder,et al.
The Laplace Transform
,
1943
.
[6]
A. M. Hasofer.
ON THE INTEGRABILITY, CONTINUITY AND DIFFERENTIABILITY OF A FAMILY OF FUNCTIONS INTRODUCED BY L. TAKACS
,
1963
.