By Luis Lopez, AI transportation consultant, CEO of Go Hub.io Holdings Corp and subsidiaries, and host of the Freight Guru Podcast
Every shipper and broker who hires motor carriers eventually ends up looking at federal safety data. The problem is that the data is easy to find and easy to misread. A single number pulled out of context can make a solid carrier look risky or a risky carrier look fine. This article explains how the Federal Motor Carrier Safety Administration organizes carrier safety information, what the public can actually see, and how to use it sensibly when choosing carriers.
What CSA Is
CSA stands for Compliance, Safety, Accountability. It is FMCSA’s program for identifying motor carriers with safety problems and prioritizing them for intervention. The data feeding CSA comes mainly from roadside inspections, crash reports and investigations. FMCSA’s Safety Measurement System, or SMS, organizes that data so the agency can compare carriers.
FMCSA has been revising its SMS methodology. Before relying on any specific scoring approach, check FMCSA’s website for the methodology currently in use.
The Seven BASICs
CSA groups safety data into seven Behavior Analysis and Safety Improvement Categories, known as BASICs:
- Unsafe Driving. Dangerous driving behavior such as speeding, improper lane changes and texting while driving.
- Crash Indicator. A carrier’s history of crash involvement, based on state-reported crash data.
- Hours-of-Service Compliance. Violations of the rules limiting driving and on-duty time, including logging problems.
- Vehicle Maintenance. Problems with brakes, lights, tires and other equipment, plus cargo securement issues.
- Controlled Substances/Alcohol. Violations involving drug and alcohol use or possession.
- Hazardous Materials Compliance. Problems with hazmat packaging, placarding, shipping papers and related requirements.
- Driver Fitness. Issues with licensing, medical qualification and other driver qualification requirements.
Each BASIC tells you something different. A carrier with clean maintenance inspections but repeated hours-of-service problems has a different risk profile than one with the reverse pattern.
What the Public Can and Cannot See
This is the point most people get wrong. Since the FAST Act of 2015, the SMS percentiles and alerts for property carriers are not displayed publicly. If you are hiring a carrier to haul general freight, you will not see the percentile rankings that FMCSA uses internally.
What remains public is the underlying inspection and violation data. You can generally see:
- The number of roadside inspections the carrier has had.
- Violations found during those inspections, grouped by category.
- Out-of-service results for vehicles and drivers.
- Reported crash involvement.
- The carrier’s safety rating, if it has one.
So the job for a shipper or broker is not to read a score. It is to read the raw record and understand what it suggests.
Inspections and Out-of-Service Orders
When an inspector finds a serious violation at roadside, the driver or vehicle may be placed out of service. That means the truck cannot move, or the driver cannot drive, until the problem is fixed. Out-of-service results are among the more meaningful signals in the public data because they reflect conditions serious enough to stop the vehicle on the spot.
Look at the out-of-service results relative to how many inspections the carrier has had. A large fleet will naturally have more inspections and more violations in absolute terms than a small one. The rate matters more than the raw count.
Crash data
Crash records show involvement, not necessarily fault. A carrier’s truck may appear in a crash report even if another driver caused the crash. Treat crash data as one piece of context, not a verdict.
Safety Ratings
FMCSA safety ratings come from compliance reviews and investigations. There are three ratings: Satisfactory, Conditional and Unsatisfactory. Many carriers, however, are unrated. That simply means FMCSA has not conducted the type of review that produces a rating. It is common, especially among smaller and newer carriers, and unrated by itself is not a negative finding.
A Conditional rating means FMCSA found safety management controls that were not adequate in some respect. It deserves a closer look and a conversation with the carrier about what has changed since. An Unsatisfactory rating is a serious red flag.
Read Trends, Not a Single Number
As an AI transportation consultant, the first thing I look at in a carrier’s safety record is direction. Is the carrier getting better or worse over time? A few useful habits:
- Compare periods. Look at recent inspections against older ones. Improvement after a bad stretch can indicate real corrective action.
- Look at violation types. Repeated violations of the same kind suggest a systemic problem rather than bad luck.
- Consider fleet size. Small carriers have few inspections, so one bad inspection swings the picture heavily. Large carriers have more data, which makes patterns more reliable.
- Weigh severity. A brake violation that led to an out-of-service order matters more than a minor paperwork issue.
AI tools can help here by pulling inspection histories across many carriers, summarizing violation patterns and flagging sudden changes. They are useful for screening at scale. They should not replace a person reviewing the details on any carrier you plan to rely on.
DataQs Challenges
Carriers can challenge inspection and crash data they believe is inaccurate through FMCSA’s DataQs system. A carrier might request a review if an inspection was recorded under the wrong carrier, a violation was cited incorrectly or a crash should not be counted against it. When you see unusual entries in a carrier’s record, it is reasonable to ask whether the carrier has filed any DataQs requests and what the outcome was.
Using Safety Data in Carrier Selection
Safety data is one input among several. It belongs alongside operating authority checks, insurance verification, identity and fraud screening and your own service experience with the carrier. Our guide on AI freight fraud detection and carrier vetting covers the identity side, and our article on negligent selection claims against freight brokers explains why a documented, consistent vetting process matters.
A few principles keep you from overreaching:
- Write a policy. Decide in advance what you check and what disqualifies a carrier, then apply it consistently.
- Do not invent your own scores. Avoid creating pseudo-percentiles from public data and presenting them as official measures.
- Document your review. Record what you looked at and when, so you can show your process later.
- Re-check periodically. A carrier’s record changes. Monitoring matters as much as the initial check.
Read carefully, a carrier’s safety record tells a story about how it maintains equipment, manages drivers and responds to problems. Your goal is to understand that story, not to reduce it to one number.
For more conversations on freight and AI, subscribe to the Freight Guru Podcast.
About the author: Luis Lopez is a Miami-based AI transportation consultant and logistics entrepreneur, the CEO of Go Hub.io Holdings Corp and subsidiaries, and host of the Freight Guru Podcast.