When Dr. Haitao Li talks about optimization, he isn’t thinking only in equations. He’s thinking about barges on the Mississippi, vertical farms on city rooftops, and sensors tracking food safety across a continent. 

A professor and chair of Supply Chain & Analytics at the University of Missouri-St. Louis, and founder of the university’s Laboratory of Advanced Supply Chain Analytics (LASCA), Li focuses on how data and decision science translate into practical systems that shape how goods move, industries operate, and communities thrive. He is also the author of the 2023 book Optimization Modeling for Supply Chain Applications, published by World Scientific.

Li’s research has been supported by the National Science Foundation, the U.S. Army Research Office, the U.S. Department of Transportation, and several industry partners including Ameren, Express Scripts, HP Labs, and Cass Information Systems. He also holds multiple patents and has an extensive publication record across supply chain optimization, analytics, and decision science.

From Transportation to Vertical Farming

“Our focus is primarily on real-world relevant applications in the broad domain of supply chain, especially the use of optimization which is really my research focus” Li says.

Much of that work begins close to home. The St. Louis region, with its unique position at the confluence of two major rivers and six Class 1 railroads offers a living laboratory for transportation and logistics research. Li’s team studies how to increase the use of river transportation in combination with rail and trucking. 

“River barge transportation cannot work by itself. The question is how we design and plan from the business side to show its potential advantages and achieve efficiency and sustainability in the long run,” he explains.

Beyond transport, Li’s group explores advanced manufacturing – how sensors and the Internet of Things can enable more connected, distributed production systems. 

“With facilities and companies now linked in real time, you have more autonomous decision-making units. But at the system level we still need coordination to achieve overall performance,” he says.

That same systems thinking guides his research on food and agriculture supply chains. Supported by the Missouri Agricultural and Small Business Development Authority, Li’s team studied the economics of vertical farming and why so many North American startups fail when they try to scale. 

“Vertical farming has clear advantages: year-round availability, shorter transport, high yield, but the economic viability is often the missing piece. The gap we saw is that precision agriculture focuses on production, not on the larger supply-chain design,” he says.

The lab built optimization models to examine crop portfolios, market demand, pricing, and location decisions for urban farms, helping startups plan what to grow, when to grow it, and how to serve their markets sustainably.

Data, Collaboration, and Caution with AI

Across all these domains, geospatial data plays a defining role. For Li, location intelligence is essential to designing better supply networks. 

“We are talking about the basic geospatial data – where production facilities, warehouses, and retail stores are, and the demographic data around them. But also novel data, like satellite imagery of port congestion or the movement of ocean containers,” he says.

One of his recent research models alternative shipping routes that would bring more Asian imports through Gulf ports and up the Mississippi. “It means longer lead times, so we need data to feed optimization models that plan procurement and shipments ahead,” he explains.

That same data-driven approach underpins a new Food Chain Performance Lab spearheaded by the Supply Chain Analytics Center of Excellence at UMSL, aimed at breaking data silos between players in the food industry ranging from the input sector to producers and processors. “It’s all about data sharing. Without data, no matter how fancy the analytical method, you can’t build impactful solutions.”

Li is quick to acknowledge the private-sector giants such as Amazon, UPS, and others, that have taken optimization to extraordinary scales, but sees them less as competition than as proof of concept. “They are raising the bar, but industry still needs academia. Companies focus on immediate ROI; universities can take on exploratory, high-risk problems. The best progress happens when we work together,” he says.

Generative AI, Li thinks, is “a more powerful search engine,” comparable in impact to the arrival of Google for researchers two decades ago, but not a replacement for human judgment. 

“I don’t think we should allow generative AI to do the task for us. It can enhance productivity if used in the right way, but it also makes things up. We need validation, and that requires knowledge,” he says. 

Li uses AI tools for coding examples and classroom demonstrations but remains wary of bias and ethical risks. “Use it cautiously. It’s exciting technology, but it needs human oversight.”

Building Skills and Seeing the Bigger Picture

As chair of a department that spans undergraduate through PhD programs, Li is focused on cultivating professionals who can merge analytics and domain expertise. “Our vision is to integrate data analytics within the domain of supply chain,” he says. 

The program draws both analytically trained students, from engineers, computer scientists, and economists, to working professionals in logistics and manufacturing. “The synergy between the two is the key. And experiential learning with guest speakers, workshops, class projects gives students real-world experience before they graduate.”

Perhaps his most vivid current example of geospatial data in action is the National Science Foundation’s Convergence Accelerator project to detect and mitigate salmonella contamination in the poultry industry. Working with 19 researchers and over 15 graduate students, Li’s team is developing pathogen-detection sensors and mapping tools that combine CDC, USDA FSIS, and social-vulnerability data. 

“One of our heat maps shows salmonella incidents together with social vulnerability. It’s very powerful – the mapping and visualization go beyond description to make the data accessible for decision-makers,” he explains. 

The project, in its second year under the banner ‘sensitivity.ai’, demonstrates how analytics and geospatial intelligence can converge to protect public health.

Asked what he hopes to see next, Li points to the same guiding principle behind all his work: real-world relevance and impact. “There are tons of data out there. The key is to apply it where it makes a difference: to work with partners, test ideas, and show how analytics and geospatial technology can deliver impact.”

For Li, that commitment extends to the classroom as well. His two-fold goal as an educator is to disseminate knowledge in supply chain and analytics, and to cultivate students’ ability to acquire and apply knowledge throughout their careers.