Innovative Projects Realized

Dimonoff : Prédiction dynamique d’occupation d’espace de stationnement en milieu urbain

D’ici les trente prochaines années, il est anticipé que les deux tiers de la population mondiale résideront en milieu urbain. Cette statistique presse les villes à organiser leurs infrastructures dans une logique sans précédent afin de conserver ou d’améliorer la qualité de vie des citoyens. La rapidité avec laquelle la population croît et l’indisponibilité des informations de mobilité aux gestionnaires de stationnement et automobilistes causent d’innombrables problématiques de gestion. Dimonoff en tant que chef de file mondial dans le domaine de l’IoT relié aux technologies urbaines désire
contribuer à fournir des solutions. C’est dans ce contexte que la solution Mobilité a été lancée, basée sur une plateforme IoT complète et de capteurs connectés. Cette solution Mobilité permet une meilleure connaissance des données de mobilité que ce soit pour trouver des stationnements de façon plus rapide, surveiller et contrôler à distance les différents espaces de stationnement ou encore un meilleur suivi des accès aux zones restreintes.
Dans ce contexte, nous cherchons à résoudre une problématique importante qui réside dans la difficulté de prédire l’occupation des espaces de stationnement. Les modèles de prédictions actuels sont simplistes et requièrent souvent l’intervention humaine pour la prise de décision. L’objectif est donc de développer un modèle de prédiction plus dynamique, capable de s’adapter aux différents facteurs environnementaux, géographiques, démographiques, sociaux et économiques. La plateforme contient des données d’occupations de milliers d’espaces de stationnements situés à Québec et Montréal. Les données d’occupation proviennent de capteurs individuels et/ou de barrières d’accès. Autres que les informations d’occupations, nous y trouvons des données telles que la localisation, la météorologie et des données d’accès tels les types d’utilisateurs.

Extraction of oil from sea buckthorn berries

Baylis Farms (Baylis), located in Smithfalls, has successfully planted 3,000 to 4,000 sea buckthorn shrubs on 9 acres. Looking to be a certified organic raw material supplier of sea buckthorn oil, Baylis has reached out to Seneca’s School of Biological Science and Applied Chemistry to investigate and optimize their existing pulping process and develop a cost-effective process for extraction of oil from the berry juice. This valuable natural ingredient has strong market demand in the cosmetic industry; loaded with antioxidants, the oil from the berry is an appealing ingredient for anti-aging and anti- wrinkle products. The project will consist of two primary objectives, first to improve the pulping process to obtain a cost-effective, high throughput process and second to investigate alternative oil extraction processes. The method used for the execution of the project will be a combination of specific types of known and/or newly investigated technologies. The expected project deliverables are two new processes for Baylis Farms; these combined processes will support Baylis in converting their, primary harvesting business, to include distribution of the berry oil to raw material suppliers in the cosmetic industry as well as other distributors working with naturopathic product manufacturers.

Computer Science Internship

This project will focus on the software development and service deployment of an artificial intelligence interface aimed to capture body measurements to facilitate sizing and fit for apparel production.The technology and IT industry is rapidly growing in Canada, and a lot of emphasis is laid on development in the artificial intelligence space. This project is aimed at helping reshape the retail fashion industry by enabling the use of virtual fitting rooms to assist in e-commerce sales, enhance customer experience and improve customer satisfaction rate. This project include programming, testing, debugging using tools like Kotlin and Swift and developing beta testing and reliability models in 3D APIs such as OpenGI, Vulkan and Metal. This internship will be an opportunity for Milo get hands on experience in AI product development and app development.

Data Architect Co-Op

Enterprise Data Architect Co-Op

Holding Pavemetrics : Intelligence artificielle appliquée à l’analyse de titres boursiers

Le projet vise à utiliser l’intelligence artificielle pour en venir, à terme, à analyser les données boursières pertinentes en lien avec une stratégie d’investissement de type “Qualité” et en extraire des indicateurs de cause à effet ayant un pouvoir prédictif par l’utilisation d’une approche de « Machine Learning » (« Deep Neural Network », « Data Mining » et autres). Ceci permettrait ainsi de prendre des décisions d’investissement qui minimisent le risque pour un certain rendement espéré. Les données financières et économiques de plusieurs années antérieures seront tirées d’une plateforme d’information financière complète et fiable à laquelle l’entreprise a accès.

City of Toronto Heritage Census Project

The City of Toronto wishes to assemble a comprehensive set of available historical neighbourhood-scale Census data. While a wealth of historical Canadian Census data exist in the public domain, it is largely inaccessible to users who lack specialized training and skills due fragmentation of data holdings and their storage in different formats. Making use of existing data is extremely labour-intensive, as the user must contend with multiple legacy file formats and variable naming schemes and geographic unit identification codes that are inconsistent across time. Building on established historical Census, data management, and spatial analysis expertise at Western University and the University of Toronto, this project will create a fully documented spatial dataset that contains available Census data for submunicipal areas, including uniform identifier codes for variables and geographic unit across the entire time period. While useful for the City of Toronto, the techniques used are applicable to other geographic locations and levels. The historical Census database created by this project will be used as a decision support and planning tool for the Toronto Heritage Survey. Awareness of the timing and location of historical cultural communities will facilitate engagement with both historic and contemporary community groups as the City of Toronto conducts the Toronto Heritage Survey. More generally, the historical Census database will be made available to all City of Toronto divisions to allow use of these products in a wide range of possible cases.

Sustainable Energy Production Using Biomass Gasification

Considerable growth in the global population and economy leads to an increase in fuel demand. This requires production of clean fuels with minimum CO2 emissions to maintain climate, energy, political, and social security. Hydrogen is a promising carbon-free fuel that can be produced from various renewable sources such as wind and biological systems. The microalgal hydrogen production has potentials of negative CO2 emissions and is less energy intensive, compared to the fossil fuel-based hydrogen production. Also, the operating conditions in bio-hydrogen production processes are generally atmospheric pressure and temperature. The partner organization will be benefit from this effective biohydrogen production technology toward decarbonization.

Peptide-based materials for rapid sutureless and scarless surgical repair

Every year in the United States alone, over 30 million surgical incisions are performed, and another 7 million wounds are causes by trauma. Most of those wounds will heal on their own, but nonetheless always lead to scarring no matter what kind of material is used to close the wound. There are a range of materials and techniques used to close wounds. All agree, however, that an ideal approach to wound closure should be easy to use, fast and painless, cost-effective and not create permanent scarring. Many advancements have significantly improved wound healing. However, all continue to have significant drawbacks either in how useable they are in surgery, or how well they promote healing, and all continue to leave scars. To this end, we have developed a new

Automation of a direct air capture to renewable natural gas production platform

Global warming is a leading threat to humanity. Carbon capture, especially direct air capture (DAC), is the key to its reversal. The bottleneck of the DAC expansion is its high cost vs. the insufficient value of CO2. On the other hand, Atlantic Canada has unmet natural gas demand, especially after the termination of Sable and Deep Panuke gas production in 2018. Furthermore, to address the emission challenges, the Canadian Government has put forth an aspirational mandate to blend in 10% of renewable natural gas by 2030. To tackle these challenges, Gaia Refinery is developing a platform to employ a passive wind-assisted DAC system integrated with an RNG production platform utilizing direct methanation within microbial electrosynthesis cells to facilitate the economic output of pipeline grade RNG while removing CO2 from the atmosphere. Compare with traditional organic-based RNG, our system has fewer feedstock constraints, a higher output rate, and can eliminate the need for gas purification/upgrading. We are currently developing our demonstration unit. Expertise in electrical engineering and automation will help us integrate the various modules of the system and control/monitor them through a computer interface.

Bridging the gap: understanding relationships between art, “intangible” community assets, and climate actions and outcomes

A sense that arts practices have a critical role to play in supporting sustainability is long-standing; but, how arts can contribute to addressing these challenges is a question that is becoming more urgent due to climate change. Where science has lacked effective communication about climate change impacts, there is a hope that arts can create narratives and visualizations that help form a common language and change perceptions. Meeting the challenge of sustainability will also require engaging with multiple dimensions of understanding and arts practices hold promise for exploring cultural and aesthetic dimensions, in particular. However, efforts to link arts and arts-research to climate change impacts have shed light on a critical trade-off – either arts practices retain their aesthetic-focus but lead to non-linear outcomes with poor measurement prospects, or arts practices adopt a linear and measurable process but lose their aesthetic, exploratory, and emergent core capabilities. The goal of this research project is to gather existing information and knowledge to develop a middle approach. Rather than focusing on direct links between arts practices and climate outcomes, this project will review existing literature to uncover subjective, normative, and intangible outcomes (or ‘bridging outcomes’) of arts practices that are critical for inspiring transformative action on climate change. The results of this study will contribute to the efforts of the Metcalf Foundation (Dr. Maggs) in devising climate change-related recommendations, program development, and targeted investments. These, in turn, will better position the arts sector to support sustainability and climate change-related societal goals.

Application of Machine Learning in Radiation Oncology Scheduling

The Ophthalmology department of CHUM receives almost 400 patients per day, each of which needs to go through different tests and consultations. Scheduling those appointments is a complicated task as it involves multiple shared resources, precedent constraints between tests, and uncertainty in tests’ durations, cancellations, emergency patients, etc. The current manual scheduling at the department results in high fluctuations in workload between days, inefficient use of resources, and long waiting time for patients.
In this project, we aim to propose an optimization model to schedule patients’ appointments. The problem is modelled as a multi-appointment, multi-resources patient scheduling problem. The second phase of the project aims to utilise machine learning tools to measure stochastic factors of the problem and to integrate that information into a decision-making framework to deal with uncertainty in scheduling.

WCRI SimplySMART Project

Waterloo Co-operative Residence Inc (WCRI) is Canada’s largest student housing co-operative. It has been providing co-operative student housing and programming for post-secondary students in the Waterloo region for over fifty years. As a co-operative, WCRI relies on members’ involvement in providing affordable housing and creating a supportive and welcoming community.

Project SimplySMART focuses on documenting current WCRI processes, writing content for the website, conducting market research on e-commerce and ERP software vendors, identifying necessary winning conditions for digital transformation, devising execution strategy options, and preparing a strategy implementation plan.