No label defined (Q3713833)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3941237
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3941237 |
Statements
1984
0 references
equilibrium distributions
0 references
approximate method
0 references
mixed multichain queueing networks
0 references
several classes of customers
0 references
Markov chain
0 references
steepest descent method
0 references
0.9030247
0 references
0.8961312
0 references
0.8940712
0 references
0.8887337
0 references
[For the entire collection see Zbl 0547.00055.] The paper deals with a new approximate method for the investigation of closed, open and mixed multichain queueing networks with several classes of customers and arbitrary distribution of service times. The network is described in terms of a Markov chain, which can be decomposed into ergodic subchains. The analysis starts from the equations for the time spent by customers of class r in the ith node and it is required to find mean values of customers streams leaving the nodes. A complete proof for the existence of the solution is given and the steepest descent method is proposed to find it. A special case of the network with only one class of customers is considered and it is proved that the solution is the same as the exact value of the mean intensity of the customers stream circulating through the closed symmetric network with an arbitrary number of exponential servers and arbitrary number of customers. It is shown that in this case the computing efficiency of the method is greater than that of the earlier methods; its superiority increases with the dimensions of the problem. Finally, valuable comparisons are given between the results obtained by the approximate method and the exact method for the closed network with three exponential servers, and comparisons between the results of the approximate method and simulation for the nonexponential example.
0 references