Hyundai is taking action to improve their safety data analytics systems including machine learning and artificial intelligence driving them towards a world class safety focused organization. Students on the Hyundai team will evaluate existing recall data and develop models that provide an early warning system for potential developing trends, indicating variables/areas that should receive special attention.
Hyundai Motor Company is committed to the safety and quality of its products. One of the largest issues facing Hyundai, and many other organizations, is the efficient analysis of data and converting the data into action. For manufacturers, the most critical area is safety; and ensuring issues are quickly detected, defined, and, when appropriate, remedied in the field. According to the National Highway Traffic Safety Administration (NHTSA), in 2020, there were 786 vehicle recalls affecting over 31,000,000 vehicles. These recalls represent potential safety concerns for consumers and are costly to manufacturers. Hyundai is taking action to improve their safety data analytics systems including machine learning and artificial intelligence driving them towards a world-class safety focused organization.
Early recognition and identification of problems in the field will help Hyundai take quick field action, minimizing impact to customers and dealers. Early detection and remedy leads to fewer potential safety concerns on the roadways, minimizing the impact to brand image avoids scrutiny by federal agencies. However, accurate recognition of a small number of complaints as a trend that would later become a problem is very challenging. Hyundai maintains large, detailed data sets documenting many aspects of service and warranty issues that could result in potential safety hazards. These include information such as:
- Service and repair records of individual vehicles, including:
- How the owner described the issue.
- How the dealer describes the issue.
- What repairs were completed.
- Recall campaign information
- NHTSA data that documents complaints, investigations and recalls
The student team will first review prior case studies of recalls to characterize the first reported concerns and the development of the issue through resolution. The student team will develop models that provide an early warning system for potential developing trends and indicate variables/areas that should receive special attention.
According to the National Highway Traffic Safety Administration (NHTSA), in 2020, there were 786 vehicle recall campaigns affecting over 31,000,000 vehicles. Decreasing recall campaigns lowers the safety risk to consumers and decreases regulatory scrutiny.
Data Analysis and Predictive Analytics (2-3 Students)
Specific Skills: Data analysis, modeling, data mining and machine learning techniques, data set cleaning and integration.
Likely Majors: DATA, CS, STATS
Natural Language Programming (1-2 Students)
Specific Skills: Intermediate skills in NLP.
Likely Majors: CS, DATA
Automotive Knowledge (1-2 Students)
Specific Skills: Automotive Systems knowledge. Knowledge of automotive industry and automotive safety.
NOTE: Must also have good quantitative, data, and/or coding skills and a wiliness to develop skills.
Likely Majors: Any degree + data skills
Director, Safety Analysis and Emerging Issues
Josh supports the North America Chief Safety Officer and his staff in working cross functionally within HMNA and with external stakeholders to execute Hyundai’s vehicle safety stratgey and help position the company as a leader in transportation safety by performing data analytics on customer field data, by safety hazard, to manage and decide opening investigations. Vedder has over 20 years of automotive experience and holds a BS in Mechanical Engineering from Colorado State University.
Dr. Bao is a human factors researcher who has led and conducted multiple, large, simulator and naturalistic-driving studies for industry and government sponsors. Her areas of expertise include the statistical analysis of crash datasets and naturalistic data, vulnerable road user safety, experimental design, algorithm development to identify driver states and movement, evaluation of driving-safety technologies, measurement of driver performance, driver decision making, and statistical and stochastic modeling techniques. She has been invited to give multiple keynote speeches and served on expert panels at different conferences or meetings. She has also made technical presentations on scientific project results at many international conferences with a wide range of audiences. Dr. Bao is the author of the recent IEEE e-learning course of ‘Human Factors in Automated Vehicles”.
- This project is open to all students
- International students on an F-1 visa will be required to declare part-time CPT during Winter 2023 and Fall 2023 terms.
IP/NDA: Students will sign standard University of Michigan IP/NDA documents.
Internship/Summer Opportunity: No summer activity will take place on the project.