Queueing dynamics of a mass-casualty field hospital — triage, ED, OR, ICU & ward flow under sudden surge
A field hospital is a modular, deployable medical facility — canvas or rigid-shell — built to deliver a defined package of care (triage, emergency stabilization, surgery, inpatient beds) in places where fixed infrastructure is destroyed or absent. On a normal day it runs well within its designed limits. Disaster medicine literature analyzes capacity through four resource pillars, commonly called the "4S" framework: Staff, Stuff, Space, and Systems — surge occurs when demand outpaces any one of them.
Emergency and disaster-medicine planners (Hick, Barbisch, Koenig and others) decompose "surge capacity" into four interdependent resource pools. A facility can only treat as many patients as its most constrained pillar allows:
• Staff: clinicians, nurses, surgeons, and support personnel physically present and credentialed to work. Staff is usually the tightest constraint — beds are easy to add, trained trauma surgeons are not. • Stuff: consumable and durable supplies — IV fluids, blood products, antibiotics, oxygen, surgical instruments, dressings, ventilators. • Space: physical treatment area — bays, operating theatres, ward tents, and the overflow space (parking lots, gymnasiums, tents) that can be converted under pressure. • Systems: the command, communication, transport, referral, and logistics processes that coordinate the other three — including standing plans, mutual-aid agreements, and information flow between triage and downstream beds.
A facility "at capacity" in normal operations still has some elastic reserve in each pillar; true surge planning asks how far each pillar can stretch, and for how long, before care quality degrades.
Humanitarian planners size field hospitals against expected population need using published minimum standards rather than guesswork:
• The Sphere Handbook (the humanitarian sector's core minimum-standards reference) recommends roughly 18–20 hospital beds per 10,000 population as a baseline health-system planning ratio in a stable response, with sudden-onset disaster response adding surgical/trauma capacity on top of that baseline. • WHO's Emergency Medical Teams (EMT) initiative standardizes what a deploying field hospital can promise a disaster-affected population, classified by increasing clinical capability (see the table below) — this lets a national authority coordinating dozens of incoming international teams know exactly what each unit can and cannot do before it arrives. • Average length of stay (LOS) and daily new-patient volume at baseline are the two numbers planners use with queueing math (Little's Law, introduced in Stage 2) to translate a "20-bed hospital" into an actual daily treatment throughput.
In routine operation, patients move through a simple pipeline: Triage (rapid sort) → Emergency Department bay (assessment and stabilization) → either discharge (minor cases), Ward admission (moderate cases), or Operating Room → ICU → Ward (severe/surgical cases). Each stage has a small number of staffed slots and a typical service time — minutes at triage, tens of minutes in the ED, one to two hours in the OR, and hours to days in ICU and ward beds.
At baseline arrival rates, patients rarely wait: the system has more service capacity than demand, so queues stay near zero and beds cycle in and out smoothly. The entire surge-planning problem is what happens to this same pipeline when the arrival rate suddenly multiplies — which is exactly what the next four stages walk through.
A WHO Emergency Medical Team Type 2 field hospital — just 20 beds and a single two-table operating theatre — is rated to perform roughly 7 major or 15 minor surgical procedures per day. A single overturned bus or building collapse can generate that many surgical patients in under an hour.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| EMT Type 1 | Outpatient emergency care | Triage, first aid, minor wound care, stabilization and referral; no surgery, no inpatient beds (mobile variant) | ~100 outpatients/day; fastest to deploy |
| EMT Type 2 | Inpatient surgical emergency care | ≥20 inpatient beds, 1 operating theatre (2 tables), general and orthopedic trauma surgery | ~7 major / 15 minor procedures per day |
| EMT Type 3 | Inpatient referral / complex trauma | ≥40 beds including 6–12 ICU beds, 2 operating theatres, complex and reconstructive surgery capability | Referral hub for Type 1 & 2 teams |
| Specialized cells | Adds-on to any type | Burn care, rehabilitation, blood bank, maternal/neonatal — bolted onto a base Type 1–3 team | Matches capability to disaster-specific need |
A Mass Casualty Incident (MCI) is any event that generates more patients, in less time, than the local medical system's normal resources can absorb — the definition is relative to capacity, not to an absolute casualty count. A ten-bed rural clinic can be "mass casualty" with 15 patients; a Type 3 field hospital may absorb 150 before crossing that same threshold. What matters for planning is the mismatch between arrival rate and service rate, a relationship queueing theory makes precise.
Little's Law is a deceptively simple equation from queueing theory that underlies almost every surge-capacity calculation in hospital operations:
L = λ × W
Where L = the average number of patients in the system (or a stage of it), λ (lambda) = the average arrival rate (patients per hour), and W = the average time a patient spends in that stage (wait plus service).
Applied to a field hospital: if triage normally receives 4 patients/hour (λ) and each triage assessment plus wait averages 15 minutes (W = 0.25 h), then on average L = 4 × 0.25 = 1 patient is "in" triage at any moment — comfortably inside a 3-slot triage bay.
During a surge, λ can jump 10–50× within minutes while service capacity (the number of triage slots and staff) stays fixed. Because W grows with the queue, and the queue grows with W, the two feed each other: once arrivals exceed service rate, the number waiting (and the average wait) grows without bound for as long as the surge continues — this is precisely what the next stage visualizes as the triage bottleneck.
Two of the best-documented sudden-onset disasters illustrate how far arrival rates can spike above baseline:
• 2010 Haiti earthquake: within 72 hours, field hospitals erected in and around Port-au-Prince by the ICRC, MSF, USNS Comfort, and dozens of international EMTs were each receiving waves of severely injured patients that, on peak days, exceeded 50 times an EMT Type 2 unit's rated daily surgical throughput. The Haitian health system, already fragile, lost several of its own hospitals to structural collapse — removing baseline capacity at the exact moment demand spiked.
• 2004 Indian Ocean tsunami: across 14 countries, an estimated 230,000 people died and hundreds of thousands were injured within hours of wave impact. Coastal hospitals in Aceh, Indonesia and southern Thailand were simultaneously damaged by the same event that generated their patient load — a common and dangerous pattern in sudden-onset disasters where the hazard degrades the response capacity at the same moment it multiplies the need for it.
Both events drove the modern EMT classification and coordination system: without a shared vocabulary for "what can a Type 2 team actually do," incoming international teams in 2010 and 2004 often duplicated capability in one location while leaving others uncovered.
The first thing every arriving casualty encounters is triage — rapid sorting, not treatment. The most widely used MCI method, START (Simple Triage And Rapid Treatment), sorts each patient in under 60 seconds using three checks: can they walk (Green), are they breathing after airway repositioning (if not, Black/deceased; if breathing >30/min, Red), and do they have a radial pulse and follow commands (if not, Red; if yes, Yellow).
This produces the familiar RED / YELLOW / GREEN acuity mix — commonly cited as roughly 20% Red (Immediate), 50% Yellow (Delayed), 30% Green (Minor) for blunt-trauma mass-casualty events, though the exact split varies by mechanism of injury (blast, building collapse, and burns each skew the distribution differently).
Critically, triage categorization is not static: a Yellow patient who waits too long without reassessment can deteriorate into Red, and even a Green patient can worsen if an initially-missed injury progresses. This is why re-triage at intervals — not a single sort at the door — is a core element of MCI protocols, and it is exactly the mechanism the next stage dramatizes.
In the 72 hours after the January 2010 Haiti earthquake, individual field hospitals in Port-au-Prince reported receiving well over 1,000 severely injured patients per facility per day at peak — a caseload that would take a standard 20-bed, 2-table EMT Type 2 unit more than a month to process at its rated capacity.
When arrivals sustain a rate above the triage station's processing rate, a queue forms — and, unlike a queue at a checkout counter, every minute a casualty spends waiting is a minute their physiologic reserve can be running out. This stage models the field hospital's single most dangerous failure mode: a queueing problem that becomes a clinical one, and the ethical framework — crisis standards of care — that governs how scarce triage attention gets allocated once the queue can no longer be avoided.
Basic queueing theory (the M/M/c model, with c = number of parallel triage servers) shows that as the arrival rate λ approaches the total service rate (c × μ, where μ is the rate one triage slot can process patients), the expected queue length and expected wait both rise steeply — and once λ exceeds c × μ, the queue has no stable size at all; it grows for as long as the imbalance persists, only shrinking once arrivals slow or service capacity increases.
In field-hospital terms: 2–3 triage staff each sorting a patient every few minutes gives a fixed ceiling on throughput. A mass-casualty wave arriving faster than that ceiling does not "back up a little" — it accumulates a growing backlog for the entire duration of the surge, and that backlog only clears once the response adds triage capacity (more staff, more slots) or the arrival wave itself tapers off. This is the core reason surge plans emphasize adding triage throughput immediately, rather than waiting to see if the queue self-corrects.
A patient who is provisionally Yellow or even Green at the moment of first sort is not guaranteed to stay that way. Internal bleeding, expanding hematomas, airway swelling, and shock can all progress silently over 20–60 minutes. Disaster-medicine protocols therefore call for periodic re-triage of everyone still waiting, not a one-time sort — because the population sitting in the queue is, by definition, the population that has not yet been reassessed.
This is the mechanism visualized in this stage's simulation: patients who wait too long in the triage queue are shown shifting color — green toward yellow, yellow toward red — as a simplified stand-in for real clinical deterioration under delay. It is also the strongest practical argument for over-resourcing triage relative to every other zone during the opening hour of a response: a bottleneck anywhere downstream (ED, OR, ICU) wastes time, but a bottleneck at triage actively costs lives among patients who have not yet even been categorized.
"Crisis standards of care" (formalized by the US Institute of Medicine in 2009 and refined since) describes a deliberate, pre-planned shift in the goal of care delivery when demand overwhelms even a maximally surged system. It is usually framed as three tiers:
• Conventional care: normal resources, normal standards — the Stage 1 baseline. • Contingency care: resources are stretched (e.g. one nurse covering more beds, functionally equivalent substitute supplies) but the standard of care for each individual patient is preserved. • Crisis care: absolute resource scarcity forces a shift from "the best outcome for this patient" to "the best outcome for the population of patients," including formal triage protocols that may withhold or withdraw resource-intensive interventions (like ventilators or ICU beds) from patients least likely to benefit, in order to direct them to patients more likely to survive with treatment.
Declaring crisis standards is a formal, usually government-authorized decision — not an informal drift — precisely because it changes the ethical basis of individual treatment decisions and needs legal and professional protection for the staff who must act under it.
Applying Little's Law with realistic numbers: a field hospital triaging casualties at 40 arrivals/hour but sorting only 15/hour falls behind by 25 patients every hour. After just 4 hours of sustained mismatch, 100 people are waiting to be triaged — most of them not yet reassessed since arrival.
Once red-tagged patients clear the ED, they hit the two most physically rigid bottlenecks in the entire pipeline: the operating theatre and the ICU. Unlike triage or the ED, an operating table and an ICU bed cannot be "stretched" by asking staff to move faster — a laparotomy takes the time it takes, and an ICU bed is occupied for as long as its patient needs organ support. When these zones saturate, the backlog does not stay contained downstream — it backs all the way up the pipeline.
Field-hospital and military-trauma surgery under mass-casualty load rarely aims for a single definitive operation. Instead it applies "damage-control surgery": rapid, abbreviated procedures that stop hemorrhage and contamination, close temporarily, and move the patient to ICU or ward to stabilize — with definitive repair deferred to a second operation once the patient (and the queue of other casualties) allows.
This approach trades a longer, more complete single operation for a shorter one that frees the operating table sooner — directly increasing the number of casualties who can reach a table at all during the acute phase. It is a clear, concrete example of matching clinical technique to system-level queueing constraints: the "best" operation for one patient in isolation is not always the best strategy when 40 more patients are waiting for the same table.
Two operating tables running continuously, each completing a damage-control procedure roughly every 45–90 minutes, gives a field hospital a hard physical ceiling of somewhere around 25–30 major procedures per 24-hour day even with unlimited staff rotating through — turnover time between cases (cleaning, re-draping, moving patients) eats further into that number.
For comparison, real-world ICRC-model two-team surgical field hospitals — a long-standing reference configuration in humanitarian surgery — report sustained throughput on the order of 60–90 surgical procedures per week during major deployments, reflecting realistic staff fatigue, case-mix, and turnover rather than a theoretical maximum. A single mass-casualty event generating dozens of surgical candidates in one day can exceed a week's worth of a facility's normal surgical output within hours.
"ICU boarding" describes patients who are medically ready to leave the ICU (or the OR, waiting for an ICU bed) but cannot move because the next bed in the pipeline is occupied. Because the pipeline in this simulation is Triage → ED → OR → ICU → Ward, a full Ward blocks ICU discharges, a full ICU blocks OR recovery transfers, and a blocked OR recovery bay eventually blocks the OR table itself from starting the next case.
This is why experienced surge planners watch occupancy at every zone simultaneously rather than focusing only on the zone that looks most overwhelmed: a saturated OR is frequently a symptom of a saturated ICU or Ward further downstream, not a shortage of surgeons or tables. Relieving the true bottleneck — which is exactly what Stage 5's surge response targets — is far more effective than adding capacity to a zone that is not actually the constraint.
A single field-hospital operating theatre completing one damage-control laparotomy every 45–90 minutes gives two tables a hard ceiling of roughly 25–30 major procedures in 24 hours — far below the surgical caseload a single collapsed building or bus crash can generate in under an hour.
Surge response is the deliberate act of expanding one or more of the 4S pillars (Staff, Stuff, Space, Systems) faster than the incident is generating new demand. It rarely means building a new hospital — it means rapidly converting reserve capacity that already exists nearby into usable treatment slots, and it is the difference between a queue that keeps growing and one that starts to drain.
Disaster-medicine planners deliberately separate two terms that sound interchangeable:
• Surge capacity: the ability to manage a significantly increased volume of patients who need the same kind of care the facility already provides — more beds, more staff hours, more supplies, but no new clinical skill set. Adding overflow tent beds staffed by the existing team is a capacity intervention.
• Surge capability: the ability to manage patients who need a type of care the facility does not routinely provide — specialized burn care after a building fire, mass toxic exposure, or pediatric ICU support in an adult-only facility. Capability gaps cannot be closed just by adding beds; they require bringing in new expertise, equipment, or a referral pathway to a facility that has it.
A field hospital can have excellent surge capacity and still fail patients whose needs exceed its surge capability — recognizing which kind of shortfall is occurring is the first step in deciding what kind of help to request.
WHO defines task-shifting (also called task-sharing) as the rational redistribution of clinical tasks from highly specialized health workers to health workers with shorter, more targeted training — enabling more efficient use of the human resources actually available during a crisis, rather than being limited by who normally performs a given task.
In a surging field hospital, this looks like: general-duty nurses trained on rapid protocols taking on initial triage sorting so physicians are freed for treatment; non-surgical physicians assisting on damage-control procedures under a lead surgeon's direction; and trained community health workers handling wound redressing and basic ward monitoring so nurses can be reallocated to higher-acuity beds. None of these substitutions replace specialist skill for complex decisions — they compress the bottleneck at the specific steps (like triage sorting or basic post-op monitoring) that do not strictly require it.
Two of the fastest capacity levers available in the first 48 hours of a response are physical space and bed turnover time:
• Overflow tents and converted space: erecting additional ward or ED tenting adjacent to the core structure — or converting non-clinical space like gymnasiums, schools, or parking structures into surge wards — can add a large fraction of new bed capacity within hours to two days, without needing new specialist staff for straightforward ward-level care.
• Early / accelerated discharge of stable patients: reviewing every occupied bed for patients who could safely finish recovery at a lower level of care, at home, or at a step-down facility frees the same physical bed for an incoming casualty far sooner than waiting for a "normal" length of stay. This is a Systems-pillar intervention as much as a clinical one: it requires active daily bed-management review, not just clinical judgment at the bedside.
Together, added space and faster turnover attack the ICU-boarding backup described in Stage 4 from both ends — more beds are available, and existing beds empty faster — which is exactly why the simulation shows the queue draining once these levers activate.
During the 2010 Haiti response, surgical teams using damage-control task-shifting and accelerated discharge protocols were able to substantially raise effective bed turnover, while rapidly erected overflow tenting added an estimated 30–50% more usable ward capacity within roughly 48 hours of being staged.