University of Ioannina
Θέματα θεωρίας συστημάτων εξυπηρέτησης με επαναλαμβανόμενους πελάτες
Abstract
dc:descriptionThree retrial queueing systems with the concepts of priorities and vacations are considered. In the first model there is a single server providing two phases of service in succession. Every customer receives service in the first phase and in the sequel, either proceeds immediately to the second, or joins the retrial box from where he repeats his attempt to find available the server and get a special second phase service. Whenever the server faces an empty queue, departs for a single vacation. In the second model we consider a model with a single server providing n phases of service in succession, and (n -1) types of retrial customers. Customers are placed upon arrival in a single ordinary queue. Every customer receives service in all phases and upon service completion in the i th phase, the customer either proceeds directly to the (i + 1)th phase, or joιns the ιth retrial box from where he retries independently, both from the customer in the corresponding box and from the customers from other boxes, after a random amount of time, to find a position for the (i + 1)th phase service. Moreover, every time the server faces an empty queue after a service completion epoch, departs for a single vacation. Finally, the third model accepts two lands of customers, ordinary and retrial. To start serving both types of customers, the only server needs a start up period, while when there are no ordinary customers waiting in the queue, the server performs a close down period. The ordinary customers have preemptive repeat priority, when the server performs a start up time in order to serve a retrial customer, and non-preemptive priority when the server is busy with this type of customer. During the close down period no retrial customers can access the server, while the arrival of an ordinary customer forces the server to return and prepare to serve him. Every time a close down period is successfully completed, the server departs for a single vacation. The process, describing the evolution of the first and third system is semi regenerative, by discovering a Markov renewal process on departure epochs. Using the corresponding theory, the stability conditions are investigated, while generating functions of the joint distribution of the number of customers in the ordinary queue and in the retrial box are obtained, both in transient, and in steady state using the supplementary variable technique. Performance measures are also obtained and used to derive numerical results, describing the evolution of the system. For the first model a stochastic decomposition result is also presented, while numerical results are finally obtained and used to investigate the system behavior. For the second model, the mean number of customers in the ordinary queue in steady state, is obtained in product form while the mean number of customer in each retrial box separately are also obtained, using the solution of a set of linear equations. These performance measures are used to investigate the systems performance by providing some numerical results.
Degree
thesis:*- Grantor dc:publisher
- University of Ioannina
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Dimitriou, Ioannis
- Δημητρίου, Ιωάννης
Subjects
dc:subject × 22- Συστήματα εξυπηρέτησης
- Επαναλαμβανόμενοι πελάτες
- Φάσεις εξυπηρέτησης
- Διακοπές υπαλλήλου
- Υπάλληλοι, Χρόνος προετοιμασίας
- Υπάλληλοι, Χρόνος αποφόρτισης
- Πελάτες - προτεραιότητες, Πολλές κλάσεις
- Εξυπηρετήσεις, Γενικές
- Poisson αφίξεις
- Queues
- Customers, Retrial
- Phases of service
- Server's vacations
- Start up
- Close down
- Multiclass queues - priorities
- Services, General
- Poisson arrivals
- Φυσικές Επιστήμες
- Μαθηματικά
- Natural Sciences
- Mathematics
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/23548