Operations Research II, IE 417

Insights into the Queuing Theory
Prepared by: Dr. Parisay

 
Possible applications of queuing theory 
Different 
Queuing System
  M/M/1/GD/inf/inf M/M/1/GD/C/inf M/M/S/GG/inf/inf
Given:      
arrival rate      
service rate      
number of servers      
limit on number in system, queue capacity      
       
       
Can find the following information:      
steady state condition      
effective arrival rate, those really entered to the system      
effective service rate, those being served and left system      
average number in queue      
average number in system      
average number in sever      
average waiting time in queue      
average waiting time in system      
average waiting time in server (service time)      
balk rate, lost customer rate      
server utilization      
probability that:      
    the server is idle      
    a specific server is idle      
    the system is empty      
    a customer has to wait      
    a customer has to wait more than T time duration in system      
    a customer has to wait more than T time duration in queue      
    there are less than X customers in the system      
probability of having X customers in the system      
probability of having X customers in the queue      
       
       
Can consider the following applications:      
total average cost of ………..…… per unit of time:      
    customers waiting time in system (queue)      
    servers       
    customer dissatisfaction when has to wait 
      (no matter how long)
     
    loss of customer (when balking)      
    space allocation for waiting line      
       
service level (for example):      
    queue capacity to accommodate customers 90% of times      
    probability of having more than Y customers in system is 0.1      
       
space required for line up in layout design      
space required for servers in layout design      
forecasting of required budget for equipment and space      
sever level: 
    under-utilized servers, therefore reduce the number
    over-utilized servers, therefore increase the number
       
sensitivity analysis:      
    find the best number of servers      
    find the best queue capacity for system      
    find the effect of changes in arrival rate      
    find the effect of changes in service rate      
       

Last Updated: May 1999