Innovative Projects Realized

Automated Data Labeller

All data-driven solutions today start with the ingestion (input) of data. Typically that data is messy and unlabelled. However, downstream consumers of data benefit from well-labelled data. Data labelling (assigning categories, data types, privacy and sensitivity tags, source characteristics, etc.) is usually an error-prone, time-consuming, manual effort. There are no readily available off-the-shelf tools that perform reliable data labelling today. This project aims to design and build a configurable, scalable, automated tool for classifying data fields given a data source. The automated tool will be a software product that generates label(s) for given input data sources and data fields including at least information such as entity type, entity context, and privacy/sensitivity tags, and does so using natural language processing (NLP) in conjunction with a heuristic-based expert rule system. The tool developed as part of this project will provide the partner organisation with a competitive edge in the data science/ML/AI market, helping it grow its customer base by attracting companies and organisations that do not have the technical skillset to build their own data processing systems, which in turn will lead to increased revenue for the partner organisation.

Supporting Vulnerable Populations in Emergencies

The project goal is to complete a community-wide vulnerability capacity assessment to gain a deeper understanding of the populations that require access to Emergency Social Services (ESS) during a large scale emergency, including the potential impacts of emergencies on these populations. This will aid in resource allocation and priority setting to better prepare for and meet the needs of our communities.

Anaerobic Digestion in the North – A Economic Feasibility Study

The main goals of this project are to 1. continue to assess the economic feasibility of operating an anaerobic digester in Yukon, and to 2. develop business and marketing strategies that could see the technology used at scale in the north. 3. Allow a business administration student at Yukon University to practice her business administration skills by providing opportunities to engage in the day to day activities of Balance BioGas’s operations.

Data Science Consulting and Productionization

The intern would work on a project to develop a data science product for the retail/finance sector. By working on consulting projects and with our product development team, they would work to integrate the data science processes we use into a suite of products that helps clients manage and analyze their data in real time. The project will entail tasks ranging from preliminary data exploration, visualization, and data cleaning to feature engineering and building and training prediction and propensity models for deployment.By the end of the internship, the intern will have learned and experienced end-to-end project development from preliminary data exploration and use case identification, to model development and testing and finally to product deployment in the client’s environment.

Real-time scheduling and performance evaluation tool for house erection

Traditional stick-built can take weeks to reach the lockup stage of a single house and leaves the house exposed for long time, while a single home can be built by using modular construction within a few days, which is much quicker. In addition, using advanced technology to achieve automated and standardized operations is the target within a business. Traditional building construction may have limited capacity to explore such advances approaches, however, modular construction companies are exceptions since the products are built in the facilities. Many existing research emphasize the operations in the facilities while limited research explores on site assembly part. ACQBuilt, as an advanced panelized home construction company, is approaching a highly automated and standardized operations for building homes. The project is targeting to help ACQBuilt to explore on-site operations by developing systematic time data collection platform, setting up standard operation procedures, optimizing crew management strategies and providing timely feedback to the crews to improve their productivity. The ultimate goal of the project is to develop a real-time scheduling and performance evaluation tool for site erection and back framing process.

Pathway to comparative data analytics

Linamar uses sensors and gauges to monitor machine performance in factories across the world. However, the infrastructure to fully exploit such data to improve productivity and performance is currently limited as there is no integrated environment available to uniformly collect, manipulate and analyze the data. The primary objective of this project is to create an environment for Linamar and future interns to build and test tools that address key operations related to benchmarking, sustainability, influence, and comparison. With these endpoints in mind, the intern will work in concert with LMMS to 1. Build the development environment and establish data parameters to be collected from machine sensors and gauges; 2. Develop insightful visualizations of the data to establish benchmark data; and 3. Provide guidance on potential insights data could provide as well as limitations, and lay the groundwork for building advanced but practical analytical tools that create value to Linamar.

CURating Clinical Trials: Helping patients find hope by exploring clinical trials opportunities

The CURating Clinical Trials project will fill the service gap to help patients, their families, or clinicians find potential clinical trials for patients. Patients can apply for our service through our website and link directly with a navigator to support their search for possible trial matches.
Through the development of a new engagement platform and in partnership with an information technology company we will be able to automate some of the navigator services and increase the number of patients we can serve. Website development and wireframe development for this support is ongoing. The website should be fully functional in January 2023. Through this proposed project, we plan to determine the effectiveness of this Clinical Trials Navigator program and improve patient access to clinical trials in Ontario by helping patients find clinical trials outside of their own centre. This will also support the expanded access of the Clinical Trials Navigator program across Canada

Evaluating the effectiveness of a digital resilience training & developing the BESAFE at work scale

AIR Institutes is launching a new resilience online training through a digital portal called AIR Space. The goal of the proposed project will be to evaluate if the training delivered in an online format is effective in developing and improving resilience. As users go through different activities delivered through the digital portal, their level of resilience will be measured over time through short surveys and daily check-ins. This will allow to identify any improvement in resilience and whether the digital portal is effective in improving resilience. Additionally, AIR Institutes are also interested in developing a psychological scale intended to measure psychological safety, which is a topic closely related to resilience.

New Methods for the Synthesis of mRNA Cap Analogues

mRNA is an essential biomolecule that is ‘translated’ into protein which is another type of biolmolecule that is essential to life. All mRNA has what is called a cap. The cap portion of the mRNA is necessary for initiating the translation process. To make the capped mRNA for therapeutic applications, such as mRNA vaccines, the cap must first be prepared chemically. The industrial partner synthesizes different types of caps and attaches them to mRNA. These capped mRNAs are then evaluated as novel mRNA therapeutics such as mRNA vaccines. The industrial partner has been using traditional approaches for cap synthesis; however, there are limitations to this methodology such as multiple and challenging purification steps, the formation of side products and low yields. The academic partner will develop new and more efficient synthetic routes to existing and novel cap analogs. This will enable the industrial partner to prepare the capped mRNA therapeutics more efficiently and at a lower cost.

State of health estimation with federated learning for battery-based energy storage systems

We have aspired for a green and intelligent future, where humans, the built environment, and nature are interconnected as a cyber-physical system. To such an Internet of Things, the sustainability and robustness of the power system are crucial, and the reliable operation of the battery-based energy storage systems is essential because of their abilities in power smoothing and shifting. However, estimating the state of health (SOH) for batteries is challenging due to privacy concerns. In this project, we aim to design a SOH estimation scheme that enables multiple energy storage systems to build a common, robust estimation model without sharing measurement data, thus addressing critical issues such as data privacy and data migration cost. Increasing the research into the SOH estimation of batteries will directly guide KORE Power to replace the battery in time and reduce economic losses. This research will support KORE Power’s position as a leader in the technology industry in North America and Canada’s position as one of the pioneer countries in clean energy industries.

Extended Reach Power-by-Light

Power is typically transmitted as electricity in copper wires. This works well in many cases, but there’s a major drawback: electromagnetic interference can disrupt the power supply, which is detrimental for sensitive electronic devices. Another approach is to transmit power as light, which is immune to this interference. A laser converts electricity to light, which travels through an optical fiber. At the fiber output, the light is converted back to electricity using a photonic power converter (PPC).

The most efficient PPCs work with a wavelength of light (850 nm) that loses approximately 40% of its power for every kilometer of transmission in optical fiber. Different wavelengths of light can travel further in fiber, but they aren’t compatible with established PPC technology. This project aims to develop next-generation PPCs using materials that are optimized for a different wavelength of light (1550 nm) to extend the reach of power-by-light systems. We will measure next-generation PPCs to demonstrate their performance and implement design changes to enhance PPC efficiency.

Développement d’outils numériques afin de prévenir et réduire les risques psychosociaux au travail : Une approche collaborative par un regroupement en santé durable au travail

Les milieux de travail comportent souvent des risques qui peuvent nuire à la santé des travailleurs tels que les risques chimiques ou biologiques et les risques psychosociaux (RPS). Ces derniers correspondent à des mauvaises conditions de travail (ex. : grande charge de travail, manque de reconnaissance, conflits, etc.) et peuvent amener au développement d’atteintes à la santé physique et mentale. Ces risques doivent être connus et mesurés afin de les prévenir et de les réduire. Les milieux de travails ont besoin d’outils performants pour ce faire. Le projet actuel est donc de soutenir le développement d’outils numériques pour évaluer et agir efficacement sur les RPS, dans une démarche collaborative rassemblant différents partenaires. Le regroupement sera composé de chercheurs, citoyens, représentants du domaine de la santé au travail et de la santé publique, du développement économique et de l’industrie, des cliniciens ainsi que des experts de l’éthique et du domaine des technologies de l’information.