Project: #373 Towards Data-Driven and Continuous Safety Inspection of Commercial Trucks and Trailers Progress Report - Reporting Period Ending: Sept. 30, 2021 Principal Investigator: Pingbo Tang Status: Active Start Date: July 1, 2021 End Date: June 30, 2022 Research Type: Applied Grant Type: Research Grant Program: FAST Act - Mobility National (2016 - 2022) Grant Cycle: 2021 Mobility UTC Progress Report (Last Updated: Sept. 28, 2021, 5:03 a.m.) % Project Completed to Date: 25 % Grant Award Expended: 10 % Match Expended & Document: 25 USDOT Requirements Accomplishments - What are the major goals and objectives of the project? The overall goal is to enable predictive management of commercial vehicle fleets (tractors and trailers) for balancing safety, operating cost, workforce performance through both real-time and historical inspection data analytics of commercial vehicle fleets. - Build a data analytics architecture to hold commercial vehicle tractor and trailer inspection data.  - Tie real-time data to the database connecting sensory information by use of telematics.  - Automatic identification of components that need preventive inspection and maintenance based on integrated analysis of real-time telematics data and historical inspection data. - Quantify the value of using real-time sensor data for automatic inspection for justifying automatic inspection results and identifying policy implications of the use of telematics in practice. - What was accomplished under these goals? Since July 2021, the project team focuses on investigating multiple data sources related to commercial vehicle fleet performance monitoring and has started developing algorithms for integrating the data sources identified and analyzing deterioration rates of vehicles based on historical inspection data. - Conducted extensive search of various data sources out there that show the results of inspecting tractors and trailers - Collected inspection manuals, inspection data, and observe inspection processes at Hunter Truck – Butler through the collaboration with Compuspections - Collected a number of vehicle inspection data sources, including FMCSA data for monitoring performance of motor carriers, historical vehicle inspection data accumulated by Compusinspections (an industry collaborator and deployment partner of the project). We are in the process of acquiring vehicle emission inspection and registration data from PennDOT. - Collected some real-time sensor data generated by typical Telematics systems deployed by Clarience Technologies (an industry collaborator and deployment partner of the project) for getting prepared for the integrated analysis of historical inspection data and real-time sensor data for monitoring vehicle performance - Developed algorithms for matching the records from the inspection data collected by Compuspections and the records from FMCSA data sources and generating statistical analysis results to show how the historical inspection data capture the factors that influence the failure rates of various components of tractors and trailers - Started developing text analysis and natural language processing algorithms that can automatically organize the inspection reports' descriptions of vehicles' conditions into tables for enabling more detailed analysis of inspection reports - What opportunities for training and professional development has the project provided? - Two Ph.D. students in the Department of Civil and Environmental Engineering have got the opportunity of learning the practice of commercial vehicle fleet inspection and maintenance planning, accumulating data analysis results for future presentations and research publications. - The two Ph.D. students have weekly meetings with industry collaborators to develop their skills of presenting the work to industry professionals and identifying scientific problems from practical problems, they also get feedback from the industry professionals about various data quality issues in the inspection data and have started developing new scientific methods for data imputation and new data analysis methods that enable vehicle data analytics and predictive fleet management based on partial and faulty data - How have the results been disseminated? If so, in what way/s? - Two Ph.D. students are starting the project. The current distribution of the results are through weekly meetings with industry professionals for showing them the results of data collection and analysis for supporting predictive commercial vehicle fleet management. - The project team plans to submit some conference papers around December 2021 and early 2022, and a journal manuscript around February 2022 - What do you plan to do during the next reporting period to accomplish the goals and objectives? - Continue developing a data analytics architecture to hold commercial vehicle tractor and trailer inspection data - a special focus will be on developing algorithms that can analyze inspection reports for generating an integrated database enable the query of relationships between various factors that influence the safety performance, failure rates of specific components of commercial vehicles - Identify critical vehicle components based on inspection data analysis for justifying the real-time data collection for monitoring those components using Telematics systems - Collect real-time telematics data and interview Telematics professionals for integrating the real-time sensor data into the data analytics framework - Examine sequential-decision analysis methods for quantifying the costs and benefits of inspection and maintenance actions for the overall performance of commercial vehicle fleets and predictive fleet management decision making - (Optional for the strategic long-term growth of the project)A future extension of the work could be to develop reinforcement learning methods that support the optimization of fleet management policies with principles of deciding inspection and maintenance actions in certain decision contexts (not within the scope of this project, possibly for the next-year project as a continuation of this Mobility21 project) - Approach additional industry partners who could be interested in inspection and real-time data analytics for predictive commercial vehicle fleet management - Develop conference papers and journal articles based on the research results Impacts During this reporting period, the project team started the project and have started some activities that have some impacts: - The two industry collaborators (Compuspections and Clarience Technologies) have started the process of cleaning and organizing their data for supporting integrated analysis of historical inspection reports and real-time data, and possibly start improving their software and hardware platform based on the findings of the project (e.g., critical components that deserve automatic inspection) - The project team has started developing deterioration models and algorithms for generating deterioration models of commercial vehicle components from inspection reports. These models and algorithms are contributing the scientific knowledge about vehicle deterioration models for supporting proactive fleet management, and computational methods for fusing unstructured data sources into a database for supporting queries related to identifications of critical components and suggestions of inspection and maintenance plans of certain components of certain vehicles with specific properties. Other The project team has produced the following outputs during this reporting period: - A database that connects inspection records from Compuspections and FMCSA inspection reports, forming a good basis for an integrated multi-source inspection database to be completed in the next step - Algorithms for matching inspection records from Compuspections and FMCSA inspection reports - Algorithms that generate deterioration rates of different components of commercial vehicles produced in different years, forming a good basis for algorithms that can generate deterioration models of vehicle components that satisfy certain queries (e.g., components of certain types of vehicles from certain motor carriers, or other conditions for filtering and selecting vehicles and vehicle components) Outcomes New Partners Compuspections, https://www.compuspections.com/ Clarience Technologies, https://www.clariencetechnologies.com/ Issues The Ph.D. student working on this project starts in August 2021, so that the project team expects a delay in the research expenditures and will closely monitor the situation and catch up on the research expenditure spent. Fortunately, the project's progress is still on track because the project team started working on the project in June even when the student has not yet officially started the Ph.D. program. The project had a PI change as expected but the research team still keep the core expertise to complete the proposed work.