Presence x dose x removal
Community viral load estimates the probability that at least one infectious person is present. People and activity estimate source strength. Air readiness and room volume estimate removal.
Room risk model
What is driving risk in this room?
This interactive model turns air, people, service, and context assumptions into a visible room read, a likely range, and a plain-language driver narrative.
Interactive model
A room-risk read before the operating question.
The goal here is not to prescribe work. It is to make the leading pressure visible so the next question is grounded: air, people, service, or context?
Illustrative room read
0
Routine
Likely range
0-0
High-side case
0
Driver read
The room is in a normal range.
Move a slider to see which layer becomes the leading pressure.
Next questions
- Which layer is driving the room read?
- What should change if that layer rises?
- What evidence would confirm the read?
Score contribution
Which lens is actually moving the score?
In this nonlinear model, contribution means marginal lift: how much each lens adds to the current score compared with a neutral baseline. That is more honest than forcing every lens into a fixed linear slice.
Air
0
marginal points
People
0
marginal points
Context
0
marginal points
Service
0
marginal points
Infectious presence
0%
chance one infectious person is present
Dose pressure
0%
nonlinear airborne pressure
Clean-air gap
0
CADR cfm estimate
Person-hours
0
per week
Show the work
The score is an argument, not a verdict.
The model starts with airborne-risk logic instead of a flat weighted average: likelihood that an infectious person is present, estimated aerosol dose over an exposure window, and effective clean-air removal. Service and context then add operational pressure without pretending they replace air control.
Nonlinear score
The score uses an exponential dose-response shape: risk rises slowly in ordinary conditions, then accelerates when infectious presence, crowding, activity, and weak clean-air removal combine.
Marginal drivers
The contribution chart asks what happens if one lens is returned to a neutral baseline. That is why a high-prevalence scenario can make context dominate even when the room equipment has not changed.
Formula structure
infectious presence = 1 - (1 - community prevalence) ^ peak people
dose pressure = 1 - e ^ -(presence x source strength x exposure / clean air removal)
room score = 1 - (1 - dose pressure) x (1 - context pressure) x (1 - service pressure)
- Air changes removal: ventilation, filtration condition, dust load, and room volume.
- People changes source strength and exposure: occupancy, peaks, activity, hours, and schedule variability.
- Service adds operational pressure: touchpoints, floor soil, high dust, moisture, and material issues.
- Context changes background risk and operating choices: room size, expected capacity, outdoor AQI, wildfire smoke, heat, humidity, and community viral load.
Airborne dose logic
The kernel borrows the shape of Wells-Riley style models: risk is not linear with crowding, time, and weak removal. Dose rises from source strength, exposure duration, breathing, mixing, and clean-air removal.
Presence and rebreathed air
Community viral load estimates the chance that at least one infectious person is present. CO2 and rebreathed-air methods, including Rudnick and Milton, are the right next bridge for replacing proxy assumptions with measured room data.
Clean-air removal
Ventilation, filtration, room volume, and portable clean air are treated as removal capacity. This follows the direction of ASHRAE 241, CDC/NIOSH ventilation guidance, and CADR-style filtration thinking.
Documented assumptions
The exact slider-to-risk conversions are not validated coefficients. They are visible starting assumptions meant to be reviewed, challenged, and improved by researchers, engineers, operators, and field data.
Research anchors: Wells-Riley airborne infection modeling, Rudnick and Milton on CO2 and airborne infection risk, ASHRAE 241, CDC/NIOSH ventilation guidance, and EPA IAQ guidance. This model is research-informed, not validated; it is a transparent planning model, not an infection-probability calculator.
From risk to action
The room read is useful only if it changes the next decision.
The risk model explains what is driving the room. The intervention planner translates that read into practical cadence questions for air, filtration, cleaning, inspection, and budget.
Read the driver
Use the model to see whether the leading pressure is air, people, service, or context.
Choose the operating question
Once the driver is visible, ask which intervention is most likely to improve the room.
Plan the cadence
Compare intervention frequency with budget and labor constraints in view.