Understanding Single-Channel Queuing Models and Assumptions

Queuing models are mathematical tools used to analyze systems where customers or jobs arrive, wait in a queue if necessary, and are served. A single-channel, single-phase queuing model represents the simplest system: one waiting line (queue) and one service facility.

To analyze these systems mathematically, certain assumptions about the arrival process, service process, queue discipline, and system capacity are typically made. Let's examine the given statements to determine which are commonly involved in making single-channel, single-phase queuing models.

Statement Analysis for Queuing Model Assumptions

• (A) Every arrival waits to be served regardless of the length of the time.
  This assumption means that customers arriving at the system will join the queue and wait until they are served, regardless of how long the queue is or how long they have been waiting. This is often referred to as an infinite queue capacity and no balking (leaving upon seeing the queue) or reneging (leaving after joining the queue). This is a standard assumption in basic queuing models unless stated otherwise.
• (B) Arrival are independent of preceding arrivals, but the average number of arrivals does not change over time.
  This statement describes a Poisson arrival process. 'Independent of preceding arrivals' means that the time between successive arrivals is unpredictable and random. 'Average number of arrivals does not change over time' implies that the arrival rate is constant over the period being studied. Poisson arrivals with a constant rate ($\lambda$) are a very common assumption in queuing theory, especially for models like M/M/1.
• (C) Arrival come from a finite or very small population on LIFO basis
  Basic queuing models often assume an infinite population, meaning the number of potential customers is large enough that the probability of an arrival is not affected by the number of customers already in the system. While finite population models exist, the assumption of an infinite population is standard for many basic models. Furthermore, LIFO (Last-In, First-Out) is a queue discipline, but FIFO (First-In, First-Out) is the most common and standard queue discipline assumed unless specified otherwise. So, this statement contains elements that are not standard assumptions for typical single-channel models.
• (D) Service times follow the negative expantial distribution or are constant.
  The distribution of service times is a key element of a queuing model. Assuming service times follow a negative exponential distribution is common for models like M/M/1, which simplifies the mathematical analysis. Assuming constant service times is also used in models like M/D/1 (where 'D' stands for deterministic, meaning constant). So, these are both common assumptions about service time distribution depending on the specific model being used.
• (E) The average service rate is greater than the average arrival rate.
  Let $\lambda$ be the average arrival rate and $\mu$ be the average service rate. The condition $\mu > \lambda$ (or service rate > arrival rate) is a fundamental requirement for a queuing system to reach a steady state. If $\mu \le \lambda$, the queue would grow indefinitely, and the system would be unstable. This is a crucial assumption for meaningful long-term analysis of queue performance.

Summary of Assumptions for Single-Channel Queuing Models

Based on the analysis, statements (A), (B), (D), and (E) represent standard or common assumptions made when developing and analyzing single-channel, single-phase queuing models. Statement (C) includes non-standard assumptions (finite population, LIFO) compared to the most basic and frequently taught models.

Therefore, the set of assumptions involved in making single-channel, single-phase queuing models, according to the provided options, includes (A), (B), (D), and (E).

Revision Table: Key Queuing Model Elements

Additional Information on Queuing Models

Single-channel, single-phase queuing models are the foundation of queuing theory. The most famous example is the M/M/1 model, where 'M' stands for Markovian (referring to exponential distributions for both arrivals and service times), '1' stands for a single server. Other variations exist, such as M/D/1 (Poisson arrivals, deterministic service times, 1 server) or M/G/1 (Poisson arrivals, general service time distribution, 1 server).

Understanding these core assumptions is vital because they directly impact the formulas used to calculate performance measures like average queue length, average waiting time, system utilization, etc. Deviating from these assumptions (e.g., finite queue, batch arrivals, priority service) requires different or more complex models.

The condition that the average service rate must be greater than the average arrival rate ($\mu > \lambda$) is crucial for practical application and analysis of queuing systems in a steady state. If this condition is not met, the system cannot handle the incoming load in the long run.