Adapting retail practices to the post-pandemic landscape
Graduate student internship supported by Mitacs via the Accelerate program
The pandemic has impacted consumer behavior dramatically by spending less time shopping, shopping in larger quantities and minimizing physical contact. In this project, the MRIL will be working with Alimentation Couche-Tard to understand and predict the current behavior of customers to help them adapt their retail practices to the post-pandemic world. By leveraging both historical (pre-pandemic) and present-time data (during and post-pandemic), researchers at the MRIL will develop state-of-the-art machine-learning and artificial intelligence algorithms. More specifically, novel demand forecasting, tracking and influencing methods (e.g., nudging interventions, sending app notifications to customers) will be tested in terms of prediction accuracy and their possible application to personalization purposes. The main objective is to identify key features related to purchasing behaviour that can help retailers better predict customer demand in these uncertain times.
Retail innovation lab: data science for socially responsible food choices
IVADO-funded project
Led by a multidisciplinary team, this project will apply artificial intelligence techniques to study, implement, and validate systems for guiding customers to make healthy food choices in a convenience store setting, while being cognizant of privacy concerns.
Reinforcement Learning for Retail Counterfactuals
The motivation of this project is to use reinforcement learning on the retail innovation lab in order to develop a virtual store environment that will allow us to train virtual agents according to real-world customer trajectory and purchasing behaviours. These virtual agents will allow us to perform customer-and store-centric configuration and layout experiments at scale, before deploying them in the physical store. An example of one such experiment would be the optimization of store layout that promotes healthy consumer buying behaviors.
IVADO-funded project
What isn’t in our dataset? -- understanding missing data for potential RIL users through longitudinal observation
As automated service systems enter into our society, whether the system meets users’ expectations or not plays an integral role in gaining trust in the system. As trust reduces ambiguity by forming expectations of the system’s intentions, people are more willing to adopt automated systems that they trust. However, building trust toward AI is complex as trust is dynamic, temporal, and incremental. We propose a study to conduct a qualitative, observational study using a method called the fly on the wall approach. This method of qualitative research aims to observe the behaviours of individuals and environments in as naturalistic setting as possible with minimal disruption or modification to the environment by the researcher.
Supported by an IVADO post-doctoral research grant and funding from the Ministère de l’Économie et de l’Innovation
Towards understanding users’ usage patterns of retail innovation lab
As an extension of an observational study, we aim to identify what prevents people from entering and using the RIL. More specifically, this project aims to understand and identify customers’ behaviours or rationales of why they are adopting, or not, the novel technology, from the perspective of stakeholders who have direct interactions with the RIL users.
Supported by an IVADO post-doctoral research grant and funding from the Ministère de l’Économie et de l’Innovation
Planogram Optimization
We aim to identify the effect of product location on demand. In particular, we are interested in the impact of the vertical location of the products on sales. To do so, we are running an experiment over six Couche-Tard stores. The insights of the project can provide guidelines for the practitioners on how to design the planograms and how to determine the reservation value of the spaces in the store.
