The Statistical Analysis of Semi-Markov Processes with Applications to Queuing Problems.
Abstract
Maximum-likelihood estimators and maximum-likelihood ratio tests are derived for the parameters of a semi-Markov process. The methods presented in the paper are essentially extensions of Billingsley's methods for Markov chains and Markov processes. The results obtained are applied to a semi-Markov process representing certain state-dependent queuing models. Confidence intervals are derived for single channel queues with Erlangian interarrival times and state-dependent Erlang service times in view of the possible use of this model to approximate arbitrary GI/G/1-like models in a manner similar to that suggested for GI/G/1 by Kotiah, Thompson and Waugh. Computational difficulties in computing simultaneous confidence intervals are indicated. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 30, 1970
- Accession Number
- AD0715657
Entities
People
- Tarsaim Lal Goyal
Organizations
- George Washington University