So, you want to talk queuing with the pros? Understanding the terminology used to describe the forces of your company’s waiting line system not only makes you sound impressive, it also comes in handy when analyzing the strengths and weaknesses of your queuing system.
Speaking of queuing system… let’s get right down to it.
A queuing system is generally defined as a flow system in which a commodity moves through one or more channels in order to go from one point to another. When we’re talking about waiting lines, we refer to a queuing system which consists of entry points, the actual waiting lines, service channels, and exit points. A commodity (a.k.a. your customer) enters the system (queue) at the entry point and, if no service channel (cashier or service agent) is immediately available, the commodity is essentially “moved” into a queue. Once a service channel is available, the commodity moves from the waiting line into the service channel, and having received service, leaves the system.
As you can see from just this first definition, there is an entire body of knowledge devoted to queuing systems. As you go about the process of improving and managing your own waiting line efficiency and experience, here is a list of some of the more common queuing-related terminology you’ll be exposed to:
Commodity: a discrete unit that is expected to be serviced (customers!)
Queue: waiting line where the customer (a.k.a. commodity) remains until a service channel becomes available
Queue length: the number of customers waiting for service
Discrete: refers to an individual, unique event or commodity
Finite population: refers to the limited-size customer pool that will use the service and, at times, form a line. A change in the population affects system probabilities.
Infinite population: An infinite population is large enough in relation to the service system so that changes to the population (a customer needing service or a serviced customer returning to the population) does not significantly affect the system probabilities.
System state: the average combined number of customers in queue and in service
System capacity: noted by the integer number of allowed commodities, which, if not specified, is assumed to be infinite
Typical waiting time: average time a customer has to wait before receiving service
Steady state: a queuing system that has been in use for some time and has reached a stabilized customer flow and service rate
Stochastic process: a process that describes arrival patterns of customers as being random due to an individual’s arrival being independent of the number of other customers currently in the queuing system, an individual’s arrival being independent of the previous customer, or the irregular time frames between customer arrivals
Poisson distribution: an exponential distribution that describes the probability of a discrete customer occurring with a known average rate, independent of the time since the last customer and the current system state
Server utilization: ratio of the average time spent helping customers to the duration of time in question (also known as traffic intensity, offered work load, and the busy probability for an arbitrary server)
Server idle time: average time a server spends between helping customers
Balking: leaving the system before entering the waiting line
Jockeying: moving from one queue to another
Reneging: leaving the system after having spent some time in the waiting line
With this reference list in hand, it’s time to learn how to put this terminology to use. Our recent guide, Queuing Theory 1.0, offers a practical approach to understanding general queuing systems.