In this insightful interview, Dutt Thakkar, Project Manager at Purdue University’s Dauch Center, discusses the significant impact of Kirchner’s U.S. import data on their research. He explains how the dataset enhances their understanding of global trade patterns and improves their work in supply chain management. This valuable resource has become a key tool for Purdue’s researchers and students, enabling them to uncover new insights and contribute to the advancement of economic research.
Dutt Thakkar is a Project Manager at the Dauch Center for the Management of Manufacturing Enterprises (DCMME) within Purdue University’s Mitchell E. Daniels Jr. School of Business. With expertise in global trade, supply chain management, and operations management, Dutt works closely with students and faculty on cutting-edge research projects. His work focuses on analyzing industry trends, supply chain logistics, and regional economic development, particularly using large-scale datasets like U.S. import records to drive insights and innovation in both academia and industry.
Transcript of the Interview with Dutt Thakkar: Exploring U.S. Import Data for Academic Research
[Kirchner Opening Music] From sea to trade, across the land, Kirchner’s legacy, we understand. Global data, insights wide, For research strong, with Kirchner by your side.
Niklas Vesely: Hey Dutt, thank you so much for taking the time.
Dutt Thakkar: Of course.
Niklas Vesely: Are you at Purdue University right now, on the beautiful campus?
Dutt Thakkar: Yes, I am.
Niklas Vesely: Lucky you, that’s very nice! Let me give a brief introduction so that everyone knows who I’m talking to. You are Dutt Thakkar, currently working at Purdue University. You’re at the Dauch Center for the Management of Manufacturing Enterprises, and you also work with the Global Supply Chain Management Initiative. Your main areas of focus include global trade, supply chain management, operations management, and manufacturing processes. Your research spans global trade logistics, and you also collaborate with students on further research.
Dutt Thakkar: Yes, we also do some regional economic analysis and look at the overall supply chain.
Niklas Vesely: You’re working at Purdue’s Mitchell E. Daniels Junior School of Business. Purdue bought U.S. import records from 2014 through to 2023. So, the first question that comes to mind is, with over 70 million records, this falls into the realm of big data. How do you navigate through such a large dataset?
Dutt Thakkar: Right, it was really important to look at all those imports coming into the U.S. We bought the data starting from 2014, gradually acquiring each year’s records. As you said, it’s a large dataset, definitely in the big data category. It’s not easy to work with such a massive dataset, and it can be challenging. We mainly used Python to sort and organize the data. Then we utilized MySQL Workbench, where we ran SQL queries to extract and analyze specific parts of the data.
Niklas Vesely: Wow, that means you coded everything yourself rather than using ready-made software?
Dutt Thakkar: Yes, exactly.
Niklas Vesely: That’s impressive! What kind of research have you conducted using Kirchner’s data?
Dutt Thakkar: With Kirchner’s data, we’ve been able to look at industry trends. The key data point is the HS code, which identifies the specific commodities we’re analyzing in the import trades. We used that to identify trends in imports across different industries. Specifically, we focused on industries and companies in Indiana, since Purdue is located in West Lafayette, Indiana. We also analyzed the data to track where these imports are arriving—on which U.S. coast, which regions, and even down to specific counties in Indiana.
Niklas Vesely: It’s impressive how you went from this broad dataset to such a detailed analysis. How much data was actually left after narrowing it down?
Dutt Thakkar: When we focused specifically on Indiana, we worked with about 20-30% of the data. However, we didn’t just stop at that. We also took a national view, analyzing how those industries are prominent in other regions and how goods are moved across the country.
Niklas Vesely: Were your findings statistically significant? Did you have enough data, and how did working with this dataset compare to others you’ve used?
Dutt Thakkar: There are always a lot of moving parts when analyzing trade and economics. With the data from Kirchner, we were able to look at the impact of domestic and import flows. It helped us see which industries were growing in imports and which were focusing more on domestic production. Especially with industries like EV and semiconductors, which are at the forefront of technology, the data was quite helpful.
Niklas Vesely: Can you share any student stories or experiences working with Kirchner’s data?
Dutt Thakkar: Absolutely. Many students, especially those without corporate experience, were initially overwhelmed by the scale of the dataset—70 million records is a lot! But once we focused their efforts on specific research areas, they got acquainted with the data and were excited to see the results. Many were surprised by what they learned about the imports into the U.S., things they didn’t expect.
Niklas Vesely: I imagine it must have been quite overwhelming for them when first opening those massive files!
Dutt Thakkar: Yes, definitely! Opening these files in Excel isn’t even an option; they’re too large. We had to use Power Queries and SQL databases. It was a learning experience for the students.
Niklas Vesely: How did you help students get started with such a large dataset?
Dutt Thakkar: We started by familiarizing them with HS codes, which categorize the commodities. From there, we taught them how to run queries to find specific data—grouping it by year, country, or state. It’s a step-by-step process of getting them comfortable with the data.
Niklas Vesely: It sounds like a great learning process. Have you tried forecasting with the data?
Dutt Thakkar: We haven’t done any forecasting yet, mainly due to the impact of COVID-19. The pandemic really skewed the data, making it hard to forecast trends accurately. If there had been no COVID-19, it would have been much easier to forecast over a 7-10 year period.
Niklas Vesely: That makes sense. Do you have any specific new insights or findings from the data?
Dutt Thakkar: One of the key insights was identifying which industries in Indiana saw the highest import activity. We were also able to look at how certain industries, like the EV industry, rely on imports since the U.S. doesn’t have local production for many of the raw materials.
Niklas Vesely: How does Kirchner’s data compare to other datasets you’ve worked with?
Dutt Thakkar: Kirchner’s dataset is incredibly detailed. It allows us to track shipments from the shipper’s facility to the consignee’s address, which gives us a full view of the supply chain. Organizing and analyzing such a large dataset is challenging, but it offers valuable insights.
Niklas Vesely: Any notable success stories from your students?
Dutt Thakkar: Yes, one notable success was when students used Google API to track import addresses down to the state level. Once they figured that out, they were excited to see the analysis results and gain a better understanding of regional trade.
Niklas Vesely: It’s amazing how the data allows you to go from a broad view to such specific insights. What kind of impact has Kirchner’s data had on your research?
Dutt Thakkar: The data has had a significant impact. It helped us understand the effects of COVID-19 on global trade and allowed us to analyze the shifts toward reshoring and diversifying suppliers in the U.S. It helped us identify major commodities that are being imported versus produced and moved domestically, and how we can use that information for identifying growth opportunities in different regions.
Niklas Vesely: Would you recommend Kirchner’s data to other researchers?
Dutt Thakkar: Absolutely. Kirchner’s data is comprehensive, allowing us to track commodities from origin to destination. It offers detailed insights down to monthly trends, which is very useful for understanding supply chain dynamics.
Niklas Vesely: Thank you so much, Dutt. It must be wonderful to be at Purdue right now!
Dutt Thakkar: Yes, though it’s starting to get cold! We’re preparing for that.
Niklas Vesely: I really enjoyed this conversation, Dutt. Thanks again for your time!
Dutt Thakkar: Thank you, Niklas. We are happy to provide our feedback and help Kirchner.
Leave a Reply