What Agroterrorism Planning Looks Like in Food Production
The useful questions at facility level are who can reach the product and whether anyone is watching
Food defense inside a production facility reduces to one rule a competent team can act on. Count the points where a person with no business at the product can reach it unobserved, give every one of those points a named owner, and bring the count down. The last time I walked that rule through a plant, the first entry on the list was an ingredient receiving bay standing open to the yard for the length of a delivery, with a visitor log on a clipboard inside the door that nobody had signed since spring.
Quality assurance managers and plant engineers are the people who carry this, usually at the request of public health and agriculture officials who need a plan on file. None of them are short of competence. They are short of hours, floor space, and the authority to change a receiving schedule, which is the constraint that decides what any plan can actually contain.
The mechanism that makes an access point matter has nothing to do with chemistry. It is how much product sits downstream of the point. An opening that feeds a mixer, a surge bin, or a bulk tank distributes whatever arrives there through a whole lot, while an opening at the end of a line reaches a case. A plant that batches large therefore has fewer points worth defending and a great deal more product behind each one. That is the whole of the ranking, and it is why a vulnerability assessment is an exercise in access, observation, and the quantity behind a door rather than an exercise in exotic threats.
My working life in sanitation and pathogen control runs to more than 35 years, and nearly all of it has been spent on contamination nobody intended. I came to this subject through a diploma in agro terrorism and bio security and I serve as an expert on agroterrorism and biosecurity for the United States Department of Homeland Security, and the thing I would tell anyone expecting a different kind of conversation is that the questions do not change when the intent does. Who can reach the product. For how long. Who would notice.
Here is the claim a thoughtful peer should be willing to fight me on. I think most of what a facility spends on food defense belongs in observation rather than in barriers. A lock records nothing and a camera nobody reviews records almost nothing. A receiving clerk who is actually free to stand at the door while a load comes off is worth more than either, and the reason that position keeps losing is that it shows up as labor rather than as capital. The crew working the most underfunded hour in a food plant is also the crew with the clearest view of which doors stand open at two in the morning, and almost nothing in a plan is built to collect what they already know.
Get Daniel Lynn’s stories in your inbox
Join Medium for free to get updates from this writer.
So the rule holds, and there are three conditions under which following it makes a building worse.
The first is when reducing access points moves the traffic instead of stopping it. Lock an interior door that a sanitation crew needs twenty times a night and it will be propped by the second week, which trades two supervised routes for one that nobody is accountable for. An access count will read as an improvement the whole time.
The second is when the person with access is not an outsider at all. Most realistic access in any plant belongs to people who are supposed to be there, and a plan built on perimeter arithmetic closes out with a comfortable number while leaving that entirely untouched. This is the uncomfortable part of the subject in agriculture and it is the reason the conversation stalls, because the honest version of it asks about supervision of colleagues rather than about fences.
The third is when closing a point costs the building its sightlines. A solid door in place of a glazed one, a stacked pallet in front of a window, a fence that puts a yard out of view of the office: each buys a barrier and spends the observation that would have done the detecting. A vulnerability assessment copied from a template has the same defect as lists of high touch surfaces copied out of a guidance document, which is that it directs attention at the features of some other building.
What a competent team can do with a day is modest and worth doing. Walk the receiving and ingredient addition points with the receiving crew rather than with a clipboard, and write down for each one how long it stands open and who can see it while it does. Put a name against every one of those points and tell that person they own it. Record which points cannot be closed, and why, so that the next person to read the plan understands that it was a decision rather than an oversight. I have argued before that a verification program is defined by what happens after a failed reading, and a food defense plan is defined the same way, by what happens the first time somebody reports a propped door.
The unsettled part is measurement, and it is not a small gap. Nothing in this work produces a number. Deterrence cannot be measured in a plant, the base rate of deliberate contamination is far too low for any single facility to generate evidence from its own history, and that leaves the whole program resting on judgment that nobody can audit against an outcome. Set against that, the accidental routes are measurable, most of the realistic risk in the food supply still sits with them, and every hour spent on food defense is competing for the same budget as the sanitation program that addresses them. I do not think anyone has resolved that honestly, and a plan that does not say out loud which side of it the facility chose is not finished.
A building is defended by the people who are looking at it, and that has never been the cheapest line in a plant budget.
Daniel Lynn has worked in pathogen control and disinfection science since the late 1980s.
