The transition to online learning amplified the already present disparities existent in the classroom. Many students struggled to learn from the digital content provided because they were not created in accessible formats and an unbalanced learning environment was apparent. We discovered this was due to instructors not knowing how to create digital content in an accessible manner and there was not a strong enough emphasis to educate instructors on how to create accessible content. To address this need we created an easy to use platform with clear instructions to assist and empower instructors to create digital accessible content. In turn, this substantially improves the learning experience and evens the playing field for all students.

Project Advisor: Professor Oliver Schneider

For more about our project, please visit our website and watch a short embedded video.

Floradale Feed Mill Ltd is a bulk feed supplier serving livestock and poultry farmers in Ontario, with dozens of deliveries to fulfill daily. Currently, determining truck loading configurations and delivery routes to distribute their products is a labour-intensive manual process, performed daily by an expert dispatch employee. This project’s objective is to develop an algorithm that will effectively recommend truck loading configurations and delivery routes for Floradale’s delivery fleet. This will help them realize reductions in distribution costs, manual effort, and planning time.

Project Advisors: Professors Hossein Abouee Mehrizi and Houra Mahmoudzadeh

The team would like to acknowledge the support of Floradale Feed Mill Ltd (Greg, Amber, Kirby, and George).

For more about our project, please watch a short presentation.

Tagr is a data science experimentation platform that abstracts away the mundane elements of keeping track of experiment artifacts within the data science experimentation process, allowing for a more streamlined data science workflow. Our team has developed a Python tool that will allow data scientists to easily record experimentation metadata that the workflow produces. Throughout each experiment, metadata is compiled into a store of centralized results which serve as a single source of truth for experiments. From there, Tagr defines a standard experiment taxonomy and automatically constructs a metadata catalogue for each experiment, making the data science workflow more efficient and allowing data scientists to iterate faster on projects.

Project Advisor: Professor Stan Dimitrov

For more about our project, please watch a short presentation.