AI4DA in City Planning, Environment, Transportation + Logistics
At the AI Centre for Decision Analytics, we’ll leverage AI and optimization techniques to provide solutions to help advance the communities in which we live, work and play. Every day of our life, we are using public infrastructures which are interconnected, and consuming products and services which are produced and distributed through complex networks of suppliers, carriers, producers, assemblers, wholesalers and retailers.
At a Glance
An economic powerhouse. The transportation sector in Canada contributes significantly to the Gross Domestic Product (GDP) at approximately 4.5%, which translates to $88 billion.
Job generation. It is also a major source of employment, providing jobs for nearly 1 million Canadians.
Large household expenses. Transportation ranks as the second-largest household expense after shelter, accounting for 16% of total household spending.
Environmental consequences. Transportation is the second-largest source of Canada’s greenhouse gas (GHG) emissions, accounting for approximately 25% of the total GHG emissions, with the majority coming from on-road vehicles carrying both passengers and goods.
How AI Can Help
Public Transportation Optimization. Data analysis can help optimize bus routes, train schedules and other public transportation systems to enhance efficiency and coverage. Machine learning can be used to predict demand and adjust schedules in real-time to improve service quality.
Ride-Sharing Services. Data analysis and machine learning can help ride-sharing platforms match riders and drivers efficiently, reducing waiting times and improving service. They can also optimize driver routes to save time and fuel.
Last-mile Delivery. Data analysis and machine learning can optimize delivery routes, minimizing travel distances and improving delivery times. They can also predict demand for goods, helping companies manage inventory more effectively.
Fleet Management. Data analysis can optimize the use of vehicle fleets, ensuring that they are used efficiently and maintained effectively. Machine learning can predict maintenance needs to prevent breakdowns and downtime.
Urban Planning. Data analysis and machine learning can support urban planners in making data-driven decisions regarding land use, zoning, and the placement of public facilities. This helps create more efficient and livable cities.
Decision makers are in great need of timely, resilient, safe and sustainable solutions. AI4DA will develop and communicate knowledge on the design, management, operation, and safety of networks, as well as on network technologies and environment.
Current Projects:
This project explores smart local crowdshipping (CS) as a sustainable solution for the last-mile delivery challenge in cities. Instead of relying on dedicated couriers, it leverages the existing travel patterns of everyday commuters — called local crowdshippers (LCSRs) — who can integrate parcel delivery into their regular journeys between home, work, or school. By tapping into this unused mobility capacity, the approach aims to reduce delivery costs, vehicle miles traveled, and environmental impact, while improving efficiency and customer trust.
To achieve this, the study combines urban mobility data, behavioral models, and robust optimization techniques. Using large-scale travel data, it identifies commuter corridors best suited for integrating deliveries. Random utility models help predict when and why commuters are willing to participate, balancing incentives with convenience and environmental benefits. Finally, a distributional robust optimization framework accounts for uncertainties in shipper availability, offering both risk-averse and risk-neutral models for operational decisions such as locker placement, compensation pricing, and customer assignments. The result is a reliable and scalable crowdshipping system that aligns economic efficiency with environmental sustainability.