What flags your truck for inspection: the ISS score, PRISM, and probable cause.

Not random.

The scale house didn't roll dice. Before your truck reached the on-ramp, a system had already scored you, sorted you, and made a recommendation to the officer watching traffic. Inspection is the output — and the inputs are things you control.

A machine decided you were worth a look before the officer did.

Ask ten owner-operators why they got inspected and nine will say “bad luck” or “they had a quota.” Both are wrong — and believing them is expensive, because it turns a controllable outcome into a coin flip you think you can't influence.

Roadside selection is driven by data. Every carrier has a safety profile in FMCSA's systems, built from inspections, violations, and crashes. Software reads that profile and produces a recommendation the officer sees on the screen in the scale house or the patrol car. By the time your truck crosses the scale, the recommendation is already there.

That's the good news wearing bad-news clothes. If selection were truly random, there would be nothing to do but hope. Because it's driven by your record, it's a lever. Carriers who understand the machine spend their energy on the inputs; the ones who don't spend it on resentment.

This lesson is that machine, in three parts — the ISS (the recommendation itself), the SMS / CSA score behind it, and PRISM (which wires it to your registration) — plus the loop they create, and how to break it.

ISS.

Inspect. Optional. Pass. The officer’s three-way readout.

The tool that tells the officer what to do with you is the Inspection Selection System — ISS. It reads your carrier's safety data and returns one of three recommendations.

What ISS reads is your SMS profile — your category percentiles, your inspection and violation history, and how much data FMCSA has on you. A carrier running high in the categories that matter draws an “Inspect.” A clean carrier trends toward “Pass.”

The new-carrier trap: when FMCSA has little data on you — a fresh authority, a handful of inspections — ISS often leans toward Inspect or Optional. Not because you did anything wrong, but because the system wants data. That's why new entrants so often feel singled out in year one. It isn't personal; it's the algorithm filling a thin file. The fix is the same either way: stack up clean inspections and the profile fills in on your terms.

The score.

Seven categories, weighted and ranked against your peers.

ISS is only the messenger. The score it reads is the Safety Measurement System — SMS — the engine behind the CSA program (Compliance, Safety, Accountability). Here's just enough of it to see why it drives selection.

SMS sorts every roadside violation and crash into seven BASICs — Behavior Analysis and Safety Improvement Categories:

Three things decide how much a given violation hurts:

1. 01

   Severity weight. Violations aren't equal — each carries a weight. A brake out-of-service weighs far more than a paperwork nit.

2. 02

   Time weight. Recent events weigh more and older ones decay. A violation's sting fades over roughly two years — if nothing new lands on top of it.

3. 03

   Peer comparison. Your weighted totals are normalized by exposure, then percentile-ranked against a peer group of carriers with a similar number of inspections. Higher percentile = worse. You're graded against carriers like you, not an absolute line.

We keep this short on purpose — there's a whole piece on how one inspection actually moves the number, where the weights, the decay, and the peer group get the full treatment. The only point here is that this score is the thing ISS reads.

PRISM.

Your safety record, wired to your registration.

There's one more system worth knowing, because it closes a door carriers assume stays open: PRISM — the Performance and Registration Information Systems Management program.

PRISM links a carrier's safety fitness to the vehicle registration process — the apportioned (IRP) plate. Two things flow from that link:

• Targeting. PRISM pushes carrier safety status out to the roadside, helping states identify and target the vehicles of high-risk carriers. It's part of how a flagged carrier's trucks keep meeting inspectors.

• Registration leverage. A carrier under a federal out-of-service order can be blocked at plate-renewal time. The plate becomes a pressure point: fix the safety problem, or lose the ability to register the truck.

The statutory backbone is 49 U.S.C. § 31106 (the information-systems mandate); PRISM's registration tie was expanded under federal highway law. The message for a small fleet is blunt: your safety record isn't a separate universe from your DMV paperwork. They're wired together, on purpose.

In plain sight.

Some flags don’t need an algorithm at all.

Not every inspection starts with a score. Plenty start with something an officer can see from the next lane.

Alongside the data-driven selection, these get you looked at every time:

• A visible defect — a light out, a flat or smoking tire, a leaking or obviously unsecured load (Part 393 territory).

• A moving violation — speeding, following too close, an unsafe lane change (Part 392). The stop is lawful on its own; the inspection rides along.

• Apparent fatigue or an HOS problem — a driver who looks exhausted, or an ELD/log that doesn't add up (Part 395).

• Post-crash. After a reportable crash an inspection commonly follows as a matter of course — and it lands in the same profile.

None of these care about your percentile. They're why “my scores are clean” is necessary but not sufficient: the truck and the driver still have to be right in plain sight, every trip.

The doom loop.

A bad score buys more stops, which buy a worse score. Here’s the exit.

Put the pieces together and you get the mechanism that quietly sinks small carriers — and, run in reverse, the one that lifts them.

That's the doom loop. A slipping score gets you inspected more, which gives the system more chances to catch something, which slips the score further. Left alone it compounds — more stops, more violations, higher insurance, lost loads.

Run the same wheel in reverse and it lifts you:

• Clean inspections count for you. An inspection with no violations is positive data — it's how a thin or damaged profile heals. Getting inspected isn't the enemy; getting inspected with violations is.

• Fix the categories that dominate. Brakes, tires, lights, and HOS are where most out-of-service violations live. Kill those and you cut most of the score's fuel — which is exactly what a real pre-trip and maintenance program is for.

• Dispute data that's wrong. Not every violation on your record belongs there. When a citation is mistaken or misassigned, the DataQs process exists to challenge it — and a removed violation stops dragging your percentile.

• Watch the score, not just the trucks. You can't fix a number you never look at. Monitoring your BASIC percentiles — the way the TC Safety Index surfaces them — turns the loop from something that happens to you into something you steer.

The whole reason this lesson matters: selection isn't luck, so the fix isn't luck either. It's a small, boring, repeatable set of inputs — and every one of them is something a compliance program can run for you.

Why “bad luck” is the expensive answer.

The scale house isn't a casino. It's a mirror. What it reflects — inspect, optional, or pass — is the record you've been writing all along. Which means the way out of the doom loop isn't luck or avoidance. It's the unglamorous work of writing a cleaner record, on purpose, starting with the next trip.