Graduate Student Research Seminar Day ‑ Mar 4, 2026

You are cordially invited to the Graduate Student Research Seminar of the Department of Industrial Engineering.

Date: Wednesday, Mar 4 2026
Time: 1:00 pm - 2:40 pm AST
In Person: Room I-121, Sexton Campus
Online: In-person only

(NOTE:  Students are reminded that they must attend in person if they are planning to put this toward their Seminar course requirements.)

Schedule:

Abstracts:

Testing Implicit Safety Science Assumptions in Maritime Waterway Risk Control Options: An Empirical Analysis of PAWSA Expert Judgments 

Mehdi Adeli, MASc. student

Maritime waterway safety is a critical societal concern, as accidents may result in severe human, environmental, and economic consequences. This research examines Risk Control Options (RCOs proposed within the Ports and Waterways Safety Assessment (PAWSA) framework, which relies on expert judgment to support waterway risk management in the United States. Based on an empirical dataset of 50 unique RCOs extracted from 21 PAWSA studies conducted between 2016 and 2023, the study analyzes how PAWSA experts judge risk reduction effectiveness and supporting evidence. The presentation further investigates whether these expert judgments implicitly align with foundational safety science assumptions by classifying RCOs according to responsible actors, an adapted hierarchy of controls, and risk management phases (prevention, preparedness, response, recovery, and governance). The findings provide insight into how expert-based risk assessments reflect, or diverge from, established safety science concepts, with implications for evidence-informed maritime waterway risk management.

Using Real Time Simulation to Determine an Optimal Plan for a Waiting Room Care Provider

Robert Marot, MASc. Student

The waiting room care provider (WRCP) role is responsible for performing patient re-assessments in the waiting room to comply with government guidelines. Additionally, these WRCPs can also perform a defined set of “care directives” while a patient is waiting, such as ordering long lead time diagnostic tests. Compared to conventional practice where a diagnostic test can only be approved after being seen by a physician, the recognition and ordering of long lead time tests while a patient is still in the waiting room means that tests can be completed in advance of receiving a bed, reducing the time in a bed that is dedicated to waiting for test results. This presentation reviews the current literature on how similar roles have previously been used as well as the potential of digital twins and machine learning to support their effort. The objective of the study is then proposed, which includes creating a simulation model that draws real time data from a hospital's waiting room and uses machine learning to optimize the deployment of WRCPs
 

Hospital Location Optimization Mixed-Integer Linear Programming Model for Timely Access to EVT Treatment

Borna Baradaran-Noveiri, MASc. student

Timely access to endovascular thrombectomy (EVT) is essential for improving outcomes in patients with large vessel occlusion ischemic stroke, yet substantial disparities in access persist across Canada due to geographic dispersion, uneven population density, and variation in hospital capabilities. This study examines how the configuration of Comprehensive Stroke Centers (CSCs) can be improved to reduce time to EVT treatment and increase the number of patients eligible to receive EVT. A mixed-integer linear programming model was developed using population, hospital, and travel-time data exported from the DESTINE Health platform. The model represents patient routing through Primary Stroke Centers and CSCs, incorporates time-dependent EVT eligibility decay, and applies constraints on CSC designation. Provincial-level optimizations were conducted for Ontario, Quebec, British Columbia, Alberta, Saskatchewan, Nova Scotia, Manitoba, New Brunswick, and Newfoundland and Labrador, followed by a national-level evaluation of system performance. To account for real-world considerations not captured by optimization alone, a quantitative feasibility scoring framework was applied to assess the practicality of hospital upgrades and downgrades. The model output and feasibility-adjusted configurations resulted in consistent improvements in EVT access across most provinces. At the national level, average time to EVT treatment decreased by 16.28 minutes (6.77%), while the number of EVT-eligible patients increased by 352 (5.46%). These changes were achieved through targeted CSC reconfiguration, including the strategic upgrading of 10 hospitals and the downgrading of one hospital. The findings provide a data-driven overview of how optimization and feasibility considerations can jointly inform stroke system planning in Canada.
 

Methods to Assess Disparity in Patient Outcomes Between Urban and Rural EVT Patients in Nova Scotia

Ian McCormick, MASc. student

Stroke is a leading cause of disability and the third leading cause of death in Canada, with the province of Nova Scotia having among the highest incidences. Patients with a suspected ischemic stroke are bypassed to one of nine Primary Stroke Centres (PSC) across Nova Scotia that provide thrombolysis and one Comprehensive Stroke Centre (CSC) that provides Endovascular Treatment (EVT) and thrombolysis. EVT candidates presenting at a PSC are transferred to the CSC via ground or air transport; however, delays can lead to futile transfers where patients become ineligible for EVT on arrival and may experience worse outcomes. The current study has two objectives: (1) Compare outcomes between EVT candidates who present directly at the CSC versus those transferred from PSCs, and (2) Compare outcomes between ground and air transfers. A retrospective cohort study will analyze provincial stroke and EMS registry data from 2018–2024. Logistic regression will be used to evaluate differences in functional outcomes (mRS 0–2 at discharge), adjusting for age, sex, NIHSS, comorbidities, and transfer metrics. Power analysis indicates that with 211 subjects per group (total n=422), the study has 80% power (α=0.05) to detect a 13.5% absolute difference in good outcomes, based on effect sizes from a recent meta-analysis. This research will identify disparities in access and outcomes, inform transfer protocols, and support evidence-based improvements to rural stroke care and emergency medical services in Nova Scotia.

Contact Person:
Hamid Afshari, Ph.D., P.Eng.
email: hamid.afshari@dal.ca