Yiling Zhang
University of Minnesota, MN·Engineering
Infrastructure such as building energy systems, delivery networks and public transit must be planned years ahead. Planners do this without knowing future conditions or how people will use the services.
- Fields
Engineering
Mathematics & Statistics
Computer Science
- Topics
stochastic optimization
multi-stage optimization
distributionally robust optimization
- Grants
- 2 known grants, through Aug 2029
What they work on
- Build a hierarchical stochastic optimization framework that links short-term adaptive user behavior to long-term infrastructure design, applied to building energy management and last-mile package delivery with parcel lockers.
- Develop data-driven methods for unknown uncertainty from several sources, both behavioral and contextual, and design efficient algorithms with provable guarantees on solution quality.
- Design multimodal transit systems that combine fixed-route transit with shared mobility, optimizing service planning, network design, fleet size, vehicle routing and relocation under uncertain travel times, demand and rider choices.
- Learn rider preferences from data and model demand for multimodal transit as a step-by-step adoption process, using distribution-free optimization that does not assume a known probability model.
More about the research
A building's energy use may change with the weather. Shoppers may choose different pickup spots depending on where parcel lockers are. Riders may or may not choose a new transit option. This work builds mathematical and computing tools for making long-term decisions under this kind of uncertainty. The main tool is stochastic optimization, which finds the best plan when some inputs are random. Both projects share a hierarchical approach. Long-term design choices sit at one level and short-term user responses sit at another, and each level shapes the other.
Why it matters
Better planning tools could help cities and companies build transit, delivery and building energy systems that hold up when demand and conditions change. In transit, the work aims to make combined bus-and-shared-ride services more practical. This could give car-less and underserved households affordable access to jobs, while reducing car ownership, greenhouse gas emissions and traffic congestion. Building operators and delivery services could also allocate resources more efficiently.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Yu Yang
University of Florida, FL·Engineering
Create a framework that scores the positive and negative effects of emerging systems such as shared e-scooters on different population groups with multi-level utility functions, combined by a social welfare function.
Chiwei Yan
University of California-Berkeley, CA·Engineering
Develop state-aware demand control models that use real-time vehicle and rider movement to shape demand toward ride-pooling in autonomous fleets such as robotaxis, robovans and urban air mobility.
Viswanath Nagarajan
University of Michigan, MI·Computer Science
Design approximation algorithms with provable performance guarantees for sequential testing of systems made of interconnected components, including series-parallel structures, under complex cost structures.
The school
University of Minnesota: 88 labs on PI Pages, $1.4B a year on research, 722 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.