Azhagan Avr: Powering Progress Through Data

We’re an interconnected nationwide team, but we each have our own reasons for being passionate about the work we do. This month, get to know Azy Avr, transportation data analyst in our Tallahassee office.

What’s your background and how did you become a transportation professional?

I earned my bachelor’s degree in electronics and communication engineering at Loyola-ICAM College of Engineering and Technology in Chennai, India. After graduating, I joined the Indian Institute of Technology, Madras as a project associate in the transportation department under the Intelligent Transportation Systems subdivision. That’s where I was introduced to the transportation field.

Initially, I worked on installing, repairing, and maintaining GPS devices in public buses to gather GTFS data. Then I shifted to working on Bluetooth devices for an origin-destination study. This was a unique project where I assembled and developed Bluetooth setups to collect and transmit data remotely. I designed, built, and installed five such devices along a highly congested corridor and managed both the front and back ends of the process.

When I decided to pursue higher studies in the US, I focused on universities with strong programs in both electrical engineering and transportation. That led me to North Carolina State University, where I earned my MS in electrical engineering and worked as a research assistant at the Institute for Transportation Research and Education (ITRE).

At ITRE, I worked on two major projects: pedestrian incident detection using AI and studying driver behavior using GPS data, which later became the focus of my thesis, Characterizing Driving Behavior by Using LIDAR and GPS Through Traffic Stream Models. I studied human car-following behavior using LIDAR devices and transportation models, which ultimately shaped my career as a transportation data analyst.

Outside of work, I would say I’m not a very adventurous person or a traveler. I prefer to stay at home and chill with family. I’ve played cricket since my school days and recently I was a part of the championship-winning team in the Oregon Cricket League during 2021-2022. I also love watching and talking about cricket. I love to cook, I cook different traditional South Indian dishes. We also had a stall put up last year at local community event in Tallahassee, serving Indian Chaat items.

What motivated you to pursue transportation data analytics as a career path?

With a background in electrical engineering, I’ve always enjoyed coding. While working at ITRE, I initially focused on developing the LIDAR devices for data collection. As I reviewed the quality of the collected data, I became fascinated with analyzing and visualizing it. This interest grew as I worked on my thesis, and I realized data analytics was the perfect combination of technical challenges and creativity. Being a data analyst lets me do what I love while constantly challenging me to grow.

What are a few of your favorite projects you’ve worked on so far at Kittelson?

I have so many favorite projects, each meaningful in its own way.

• • Congestion Metrics Program in Washington County, Oregon: This project is special to me because it was the first product I delivered to a client. I developed a Python-based tool to automate the entire process of downloading, analyzing, and visualizing Bluemac data for the county’s Bluetooth Analysis Tool. The automation made labor-intensive and error-prone tasks seamless, and the client was so impressed they returned to us for additional features.
  • MassDOT I-93 Bus-On-Shoulder Dashboard: This was my first web-hosted tool, and while I worked on the backend programming, merging a Python backend with a JavaScript frontend was a rewarding challenge. I developed Python code to integrate multiple data sources, including permanent count stations, vehicle probes, and transit agency data, into an intuitive dashboard. The web page updates automatically every month, providing real-time insights.
  • PennDOT WebICE Tool: I learned Java for this project, which was incredibly satisfying. I coded computation formulas for delays, costs, and safety evaluations, converting PennDOT’s Excel-based ICE tool into a web-based version. It was challenging initially, but I thrive on coding challenges, and seeing it come together was deeply rewarding.
  • NCHRP 07-26 Highway Capacity Manual Update: This was a coding-intensive project where I helped automate processes to evaluate the capacity of over 120 locations using methodologies developed for the Highway Capacity Manual. I also streamlined figure visualizations across locations. The project required integrating diverse data sources, which made data wrangling and fusion a challenge—but automation saved us countless hours.
  • Crash Recorder: I developed the Crash Recorder Tool (for Kittelson’s internal Innovation Kitchen), used in Oregon to analyze and visualize crash data. What used to take 20 hours now takes minutes. The tool has evolved from an Excel tool to a Python-based desktop app and is now a Python-based web tool. I currently manage the tool, ensuring it scales and remains effective while mentoring others to maintain it.

How do you stay up to date with new tools, technologies, and methodologies?

I’m always learning through professional certifications and online courses to stay current with industry standards and processes. For instance, when Kittelson’s software group introduced more collaborative tools, I transitioned from PyCharm to VS Code, which offers enhanced GitHub integration for better teamwork.

I also attend conferences like TRB and ITE events to network with professionals and stay informed on industry advancements. Staying current is critical for maintaining a collaborative and innovative environment.

What trends or innovations in transportation data analytics do you think will have the biggest impact in the next 5-10 years?

AI, especially generative AI, has revolutionized many aspects of engineering. Tools like ChatGPT highlight the importance of prompt engineering, where professionals must know how to ask the right questions and interpret AI-generated suggestions critically.

While there’s concern about AI replacing human roles, I believe we’re far from that reality. AI relies heavily on the quality of available data, and flawed data can lead to flawed results. The next 5-10 years will be an exciting period as we learn how humans and AI can coexist and complement each other in advancing the transportation industry.