#373 Towards Data-Driven and Continuous Safety Inspection of Commercial Trucks and Trailers

Principal Investigator
Pingbo Tang
Start Date
July 1, 2021
End Date
Aug. 30, 2022
Research Type
Grant Type
Grant Program
FAST Act - Mobility National (2016 - 2022)
Grant Cycle
2021 Mobility UTC


Currently, highway trucks (a.k.a. tractors) and trailers are subject to a patchwork of state and federal safety inspections which manually and periodically check safety components like tires, brakes, and lights either through planned or unannounced roadside inspections, which can delay deliveries and increase costs. 
We will partner with two Pennsylvania firms to create a connected truck system platform that can continuously monitor tractor-trailer safety, perform comparative analytics for supporting predictive maintenance and fleet management, and report the results of safety systems and the cost & benefits of implementing such a connected truck system to entities like state police and DOTs.
Around the world, vehicles of many types are subject to periodic inspections to assess the current status of their safety or emissions management systems. In the US, there is a complex system of rules for safety inspections in order to keep vehicles safe. These programs check vehicle safety components like brakes, tires, lights, etc. Safety inspections for passenger vehicles are typically done annually or biennially; however, there is no federal inspection requirement for passenger vehicles, and less than half of states have mandatory safety inspection programs. State legislatures in the past decade have been eliminating these programs, or making them less frequent, due to the perception that the programs waste time and money and unfounded beliefs that since vehicles have never been safer that vehicles rarely fail the inspections.
In past work, we created a substantial data analytics infrastructure to collect data on and assess the effectiveness of state safety inspection programs for passenger vehicles [Peck 2015]. This has been done through a consortium arrangement where various stakeholder institutions, including private companies and global inspection industry associations (e.g., CITA) have contributed cash and in-kind support for our ongoing work. Data provided by industry partners and PennDOT shows the results of all inspection criteria for all vehicles available, and a key finding was that contrary to the popular belief that few cars fail safety inspections, about 15% of vehicles would fail inspections had they not had maintenance done at the time of inspection to be able to pass (e.g., had tires or brakes replaced). Our most recent study shows that states with passenger vehicle safety inspection programs have fatality rates about 2% lower than states without programs [Acharya 2020]. This lends strong support to maintaining not eliminating inspection programs.
In this project, we turn our attention to the 3 million commercial trucks/tractors and 6 million trailers in the US. Commercial trucks and trailers are subject to annual federal Department of Transportation (DOT)-required inspections, and usually, separate annual state inspections and random ad hoc roadside inspections conducted by police. These safety inspections check brakes, tires, lighting, couplings, exhaust and fuel system, etc. 
As compared to passenger vehicles, there is less publicly available data for the results of commercial inspections at the vehicle level. Instead of the vehicle-level data used above for passenger vehicles, federal and state reports typically provide only aggregated summary data about the results of inspections (e.g., the total number of annual or roadside inspections done for an entire fleet and the total number of violations identified over all inspections), making it hard to reach detailed conclusions about how the current system performs [FMCSA 2020].
In addition to the issues mentioned above, commercial truck inspections consume time, which is valuable in a freight mobility environment that is increasingly constrained. The added uncertainty of random roadside inspections further adds to potential costs to truck and trailer owners. As with passenger vehicles, there have been suggestions to reduce the burden of commercial truck and trailer owners since they are inspected so frequently (as a result, Pennsylvania recently changed from semi-annual to annual truck inspections). Fleet owners also seek to be exempt from roadside inspections since they claim to have centrally managed inspection regimes that identify and fix problems before inspectors notice them.
Given the potential for loss of time and money, it would be valuable to have data-driven systems that better understood likely inspection failure modes and also provide continuous/real time ‘dashboard like’ insights into the status of safety systems on commercial tractor-trailer units. If industry, state and federal inspectors, and others, worked together it could be possible to create a system that is safer than the one in place today but which also maximizes freight mobility by reducing delays from inspections.
We believe that the combination of data analytics of past truck inspections, and an emerging truck telematics platform (Road Ready, described below) will allow us to achieve these goals.
We envision the following tasks in the one-year Mobility21-funded part of the project.
Task 1: Data Analytics Architecture for Commercial Vehicle Inspection Data 
We will collate and organize the commercial truck and trailer data from PennDOT, CompuSpections, and prospectively a major truck fleet owner, and create dashboard-like summaries of it in order to provide stakeholders with fast access to better information, especially legislators interested in modifying a tractor-trailer inspection program.  For example, we will show the most failure-prone safety components, and to see whether they differ from passenger vehicles (where most failure are brakes, tires, and lights).
In support of this effort, we will develop data analytic methods to operate on the large historical data archive to identify features such as trends associated with safety component failures (e.g., brakes, tires, or lights) as well as failure rates across make-model-year of vehicle, age, and miles driven.  We envision developing code in Python and R to accomplish this task. 
The work in this (and the next) task will be led by CEE graduate student Chenyu Yuan.
While our focus in this initial period will be on Pennsylvania-registered trucks and trailers, and those owned by a major fleet operator, in parallel, we will reach out to other organization such as the US DOT, the Federal Motor Carrier Safety Administration, and other state DOTs for additional historical data from commercial truck and trailer inspections (even if only available in aggregate). Such organizations have previously contacted us about doing studies like we have previously done for passenger vehicles.
Task 2: Linking Data Analytics to Monitoring System Design
Next we want to assist with further “automating” the process of making the safety status of key components visible to drivers and fleet owners. In this task we will merge information from Task 1 on the most common component failures with the hardware/software system design of Road Ready. For example, if the most failure-prone components of commercial vehicles are brakes, tires, and lights, we will assess how effective the Road Ready system is in identifying those failed components.
We will work with Truck-Lite to assess the reliability of their telematics system.
Task 3: Discussions with State and Federal Stakeholders
Finally, as noted above, we seek to open a dialogue with public agencies responsible for setting and managing inspections, to understand whether they are willing to discuss flexibility in the timing and practice of inspections in exchange for continuous data access. For example, if they could receive feeds of continuously monitored trucks/trailers and are able to continuously monitor the identified key failure components, then perhaps a share of vehicles in fleets could be exempted from roadside inspections. While such an outcome may be unlikely in the short term, we are interested in knowing what types of data they would expect to have access to, and for how long of a period of time, before they might be motivated to pursue such a system.
Note that we intend to submit a parallel proposal to PITA for managing and funding the collaboration between CMU and the two Pennsylvania-based companies who are our main partners. This will increase funds available, and provide cash cost share and additional time to complete the project (the PITA timeline is for 18 months not just 12 months). 
The vast majority of funding requested for this project is to support a PhD student in Civil and Environmental Engineering (Chenyu Yuan, an existing CEE MS student who has already been working with Professor Matthews in analyzing the existing safety inspection data, and has been leveraging it for other purposes including two paper submissions to the annual 2021 Transportation Research Board (TRB) conference has been identified as a potential candidate for the PhD student position).  The cost share tracking for this project will be complicated due to the multiple funding sources and partners, but we are confident there will be far more than 1:1 cost share available (if both are approved, the Mobility21 and PITA sources alone will provide most of the requisite shares for each other, and will be in cash).

Dana Peck, H. Scott Matthews, Chris Hendrickson, and Paul Fischbeck, "An Analysis of Vehicle Safety Inspection Data in Pennsylvania: Expected Failure Rates", Transportation Research Part A, Volume 78, August 2015, Pages 252–265, 2015.  
Prithvi Acharya, Laila AitBihiOuali, H. Scott Matthews, and Daniel J. Graham, “The impact of periodic passenger vehicle safety inspection programs on roadway fatalities: Evidence from U.S. states using panel data”, final report to CITA and submitted to Accident Analysis and Prevention, 2020.
Federal Motor Carrier Safety Administration (FMCSA) Data Query Website,  https://safer.fmcsa.dot.gov/query.asp, last accessed December 15, 2020.
December 2020 – This proposal submitted
January 2021 – Complementary PITA proposal submitted
July 2021 – Joint Mobility21/PITA Project Starts
Task 1 (inspection data analytics): July – December 2021
Task 2 (connections to telematics): August 2021-May 2022
Task 3 (discussions with agencies): has already begun - June 2022
June 2022 – Mobility21 Project Ends
December 2022 – PITA Project Ends

Deployment Plan
For this project, we will collaborate with the following organizations:
CompuSpections, LLC – Compuspections, based in Valenica, PA, is a small business that provides web-based inspection record management and reporting software for inspection stations (ranging from small garages to large dealerships).  We have been working with them for 5 years, including on our previous safety inspection studies of passenger vehicles, and know their data architecture well. Their system records all PA safety inspection criteria, regardless of pass or fail status, and data for some fields are comprehensive, including information such as all actual tire tread measurements (in units of 1/32 of an inch) at time of inspection, all maintenance work done to meet state inspection program requirements, final pass/fail status, etc.  We expect to be able to scavenge state-mandated safety inspection data for a significant number of trucks and trailers in the state. CompuSpections has previously provided us with a large amount of in-kind data and expertise to frame this proposed study but was limited to passenger vehicles.  Given their role in the industry, they have also helped us build relationships with other interested parties such as PennDOT, various state senators and representatives, the Pennsylvania Automotive Association (PAA), the annual IM Solutions conference, CITA (a global alliance of companies in the safety and emission inspection industry), etc.  CompuSpections will be providing in-kind support in the form of all of their historical data related to truck and trailer inspections, as well as substantial staff time in extracting, providing, and supporting our work with the data (see support letter).
Truck-Lite Co., LLC – Truck-Lite is a subsidiary of Clarience Technologies LLC, a truck safety component and system company having a global R&D center located in Pittsburgh. Truck-lite has been making lights for medium and heavy-duty trucks for over 60 years and has recently entered the telematics space for commercial trucking through its Road Ready system.  The Road Ready system (https://www.roadreadysystem.com) is a new hardware and services platform that collects data from various on board sensor components (including all of the safety components, not just lights) on trucks and trailers, and communicates the data and status information to the driver and centrally to fleet managers. For example, a typical trailer has about 20 lights, and any one of them being inoperable results in a failed inspection. Clarience Technologies, LLC is co-owned by a company that owns and manages a national fleet of 50,000 trucks that are subject to the patchwork of DOT and state inspections mentioned above. It is Truck-Lite’s intent to leverage its connections with company partners to obtain access to several years’ worth of fleet inspection data to be used in this analysis. Truck-lite will be providing in-kind support in terms of staff support time, access to the Road Ready platform, etc. (see support letter).
PA Department of Transportation (PennDOT) – We have negotiated a data usage agreement with PennDOT that allows us to purchase a complete list of all registered vehicles (including commercial trucks and trailers) currently in the state as of time of request, including vehicle identification number, zip code, county, type of vehicle, etc. We will also attempt to work with them and the Pennsylvania State Policy to identify interested parties for a pilot “fleet dashboard” of our system (status not finalized at time of proposal submission).
CITA - CITA is the international association of public and private sector organizations actively involved in mandatory road vehicle compliance. While much of their members’ activities are focused on passenger vehicle programs, there is a significant cross-section of members whose operations span commercial vehicles, and are interested in the results of this work. We are also hopeful to get a cash donation from CITA in support of this work. We have previously worked with CITA in assessing how passenger vehicle safety inspection programs lead to reduced fatality rates (a recent UTC project where CITA contributed about $50k in cash).
Expected Accomplishments and Metrics
In this one year project we will:
•	Organize and analyze historical data on the components most likely to lead to a tractor-trailer safety inspection failure from state and eventually internal corporate records
•	Help to improve the telematics effectiveness of cutting edge sensing and communications technologies to continuously monitor safety component status in trucks. 
•	Begin necessary discussions with agencies about regulatory relief for fleets incorporating advanced telematics to improve freight mobility.

In this one year project we will:
•	Organize and analyze historical data on the components most likely to lead to a tractor-trailer safety inspection failure from state and eventually internal corporate records
•	Help to improve the telematics effectiveness of cutting edge sensing and communications technologies to continuously monitor safety component status in trucks. 
•	Begin necessary discussions with agencies about regulatory relief for fleets incorporating advanced telematics to improve freight mobility.

We believe the combination of parties and technologies can lead to a new paradigm for tracking commercial vehicle safety, and more importantly, save time and money for fleet owners. 
We envision the metrics of this project to be:
•	Number of inspection records available and processed
•	Number of key safety components able to be effectively monitored by the system
•	Number of external presentations given
•	Number of state agencies in discussions about using telematics system
•	Number of papers published

Individuals Involved

Email Name Affiliation Role Position
ptang@andrew.cmu.edu Tang, Pingbo Carnegie Mellon University PI Faculty - Untenured, Tenure Track
chenyuyu@andrew.cmu.edu Yuan, Chenyu Carnegie Mellon University Other Student - PhD


Amount of UTC Funds Awarded
Total Project Budget (from all funding sources)


Type Name Uploaded
Data Management Plan Data_management_Plan_for_Mobility21_Truck_lite_2021.docx Dec. 17, 2020, 2:25 p.m.
Project Brief Project_PowerPoint_Summary.pptx March 22, 2021, 6:07 p.m.
Presentation Mobility21_Truck_Fleet_Management_August_27th.pptx Sept. 26, 2021, 5 p.m.
Progress Report 373_Progress_Report_2021-09-30 Sept. 27, 2021, 11:20 a.m.
Publication UTC_Conference_Abstract_v3.pdf March 26, 2022, 6:21 p.m.
Presentation 7th_Annual_UTC_Conference_for_the_Southeastern_Region_-_03-25-2022.pptx March 26, 2022, 6:21 p.m.
Progress Report 373_Progress_Report_2022-03-30 March 27, 2022, 6:21 p.m.

Match Sources

No match sources!


Name Type
CompuSpections Deployment Partner Deployment Partner
Truck-lite Deployment Partner Deployment Partner