The Score Function (SF) method has been proposed to estimate the gradi
ent of a performance measure with respect to some continuous parameter
s in a stochastic system. In this paper we experiment with the use of
this estimate in a stochastic approximation algorithm to perform a sin
gle-run optimization. The experiment is done on a simple M/M/1 queue.
The performance measure involves the average system time per customer
at steady-state, and the decision variable is the service rate. The op
timal solution is easy to compute analytically, which facilitates the
evaluation of the algorithm. Combined with appropriate variance reduct
ion techniques, the method has been shown to be effective in the test
problem. We study the algorithm's properties and examine the validity
of the estimates based on this single run procedure by performing some
experimental studies. Implementation ''details'' necessary for packag
ing this method with existing simulation software are provided. Finall
y, there is a set of recommended directions for future research. Copyr
ight (C) 1996 Elsevier Science Ltd