G/G/1 Queue¶
A single-server queue model with general arrival and service behavior whose waiting results depend on declared sequencing and analysis assumptions.
Core Idea¶
A G/G/1 queue is a mathematical model of arrivals, service requirements, and one server. The two Gs permit generally specified arrival and service behavior; the 1 says one resource serves at a time. An ordered queue state connects those inputs to possible waiting. A chosen service discipline and capacity rule can change the answer to a particular delay question, so the label alone is not a waiting-time formula.[^ref-85e4bb57d3dd]
Scope of Application¶
The one-server model appears in two documented settings:
- Packet output. A UC3M dissertation models a switch's packet output under a declared first-come-first-served calculation.[^ref-96a9859b00f3]
- Workstation lots. A manufacturing study models one station handling lots within a larger POLCA shop.[^ref-b45cc74b4645]
Both use arrivals and service requirements at one resource, but their carriers and numerical objectives differ. G/G/1 names the node, not the whole network or factory. G/G/2 and G/G/k permit more simultaneous servers and are neighboring models, not positive G/G/1 cases.[^ref-b45cc74b4645]
Clarity¶
The Gs do not silently mean independent, identically distributed arrivals and services. Foss treats a general stationary joint driving sequence, then considers a narrower GI/GI/1 case with additional independence assumptions.[^ref-85e4bb57d3dd] Nor does G/G/1 by itself guarantee FIFO order, an infinite buffer, a steady state, or a closed-form delay. State those conditions when using a result that needs them. A reported mean delay also depends on the driving assumptions and on whether the conditions for that particular approximation or stationary result are met.
Manages Complexity¶
The model separates what enters, what work each arrival demands, and what one server can complete. That separation makes it possible to ask whether waiting or workload grows under declared assumptions. The inputs and state rules can also be studied as formal mathematical objects without an observed packet stream or a realized wait. It also exposes which features a calculation leaves out: packet bursts, lot setup, finite space, priority, or dependence may affect a particular result.[ref-96a9859b00f3][ref-b45cc74b4645]
Abstract Reasoning¶
A Lindley recurrence can analyze waiting for a suitable FIFO sequence; a Kingman-style formula gives a conditional approximation. Neither is the definition of the whole G/G/1 family. A formal G/G/1 can also be specified without a physical packet stream or observed wait. Its current catalog placement is therefore a provisional unparented root: no reviewed live parent covers its full formal identity, and this does not claim the model is a Prime.[^ref-85e4bb57d3dd]
Knowledge Transfer¶
To recognize the pattern, identify timed arrivals, service requirements, exactly one modeled server, and an ordered waiting or workload state. Then ask which discipline, capacity, dependence, and performance question are declared. The one-server structure transfers across applications; a particular packet or workstation estimate does not transfer without its assumptions. In particular, a local FIFO convention cannot be projected onto every G/G/1 model, and a worked delay approximation cannot be reused as an exact law for different traffic or lot processes.[ref-96a9859b00f3][ref-b45cc74b4645]
Neighborhood in Abstraction Space¶
G/G/1 Queue sits in a sparse region of the domain-specific corpus (90th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2721 abstractions)
Nearest neighbors
- Cost Efficiency — 0.79
- Logic Model (Program Evaluation) — 0.79
- Program Profiling — 0.79
- Network scheduler — 0.79
- Offered load — 0.79
Computed from structural-signature embeddings · 2026-10-17
References¶
[^ref-96a9859b00f3]: Otero Pérez, Gabriel (2020). “Design and Analysis of Ultra-low Latency Fronthaul and Backhaul Networks for 5G”. Doctoral dissertation, Universidad Carlos III de Madrid, §2.7.2, printed p. 36 (physical PDF index 61, FCFS switch convention), and Chapter 4 §§4.4.2–4.5, printed pp. 62–65 (physical PDF indices 87–90). Institutional PDF. The cover says December 2020; the internal title page is dated 6 November 2020.
[^ref-b45cc74b4645]: Vandaele, Nico; Van Nieuwenhuyse, Inneke; Claerhout, Diederik; and Cremmery, Rony (2008). “Load-Based POLCA: An Integrated Material Control System for Multiproduct, Multimachine Job Shops.” Manufacturing & Service Operations Management 10(2): 181–197. DOI: 10.1287/msom.1070.0174. See §§3.2.1–3.2.2, printed pp. 186–188, and §4, printed pp. 191, 193–194. Original journal-layout PDF copy.
[^ref-85e4bb57d3dd]: Foss, Sergey (undated). “The G/G/1 queue”. Original seven-page author-hosted mathematical exposition, PDF p. 1 (§1 and opening paragraphs), with general-case discussion later in the document. Author PDF. No publication year or DOI is inferred.