Graduate Student Research Seminar Day ‑ July 16, 2025

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

Date: Wednesday, July 16, 2025
Time: 11:00 - 12:15 AST
In Person:ÌýRoom MA220, Sexton Campus

Online:Ìý°Õ±ð²¹³¾²õ

Schedule:

Abstracts:

Assessing Drone Operability and Probability of Detection for Search and Rescue Operations in Arctic Canada
Mona Hakimpanah, MASc. Student

Search and Rescue (SAR) operations in Northern Canada face significant logistical and environmental challenges, including vast remote terrain, extreme weather conditions, and limited aeronautical infrastructure. With increasing SAR demands due to climate change and growing activity in the region, Unmanned Aerial Vehicles (UAVs) are emerging as a potential tool in SAR operations. This thesis investigates the feasibility and effectiveness of UAVs in Arctic SAR, with a particular focus on multirotor UAVs due to their vertical takeoff and hovering capabilities, through two primary research questions: (RQ1) Can UAVs be used year-round in the Arctic, and where and when are they operable in Northern Canada? (RQ2) What is the Probability of Detection (PoD) of Persons in Distress (PiDs) using UAVs across different SAR scenarios? For RQ1, operability thresholds are defined based on UAV specifications and expert input, then applied to hourly weather data from Environment Canada (2018–2024) for various Northern communities. Results indicate strong seasonal and locational variability in UAVs operability, with summer months providing more favorable conditions. Some communities offer significantly higher potential for UAVs deployment than others, highlighting the importance of localized planning in UAV-supported SAR strategies. For RQ2, a fuzzy logic model is used to estimate PoD under varying conditions of visibility, weather, sensor quality, UAV speed and altitude, as well as operator expertise and fatigue. This approach accounts for the uncertainty and vagueness inherent in SAR environments and translates expert knowledge into a structured computational framework. Findings support data-driven SAR planning using UAVs across remote and weather-challenging Arctic regions.

Ìý

Towards Cellular Agriculture: An Introductory Supply Chain Model
Dawne Skinner, Ph.D. Candidate

Cellular agriculture, which uses biotechnology to produce animal-derived products, has been identified as a possible solution to reduce the negative environmental impacts associated with traditional meat and dairy production. However, available life cycle and techno-economic assessments for cultured meat production suggest that additional environmental and cost improvements are needed to compete with traditional meat production methods. The adoption of circular supply chains has been found to improve the economic and environmental outcomes of production processes. The use of agricultural and food byproducts, such as hydrolyzed soymeal, as a source of amino acids has been identified as a way to reduce cost and environmental impacts. However, the impact of these undefined sources on cell production efficiency is largely unknown. The aim of this research is to develop a novel exploratory supply chain model for a viable large-scale cellular agriculture network that considers facility location, ingredient blending, capacity design and technology selection problems. A bi-objective mixed integer linear programming model is developed to investigate the dynamics between demand, capacity design, location, ingredient blending and technology selection decisions as well as trade offs when optimizing for cost versus carbon emissions.

Ìý

Inventory Management of Perishables under Zero-Inflated Poisson Demand
Soroush Safavi, Ph.D. Candidate

Blood supply chains involve challenges of perishability, variability in demand, and costs of wastage and shortage, making the management of perishable inventory in blood supply chains particularly complex. Zero-inflation and highly variable demand patterns further complicate inventory management; however, the Zero-Inflated Poisson (ZIP) model provides a suitable framework for capturing these characteristics effectively. In this paper, we investigate how demand characteristics impact inventory performance and suggest strategies to maximize cost efficiency. This research evaluates the classical reorder-point/order-up-to-level (s,S) inventory policy under ZIP demand for a specialized product called Low-Titer O group Whole Blood (LTOWB). A ZIP model is fit to Canadian demand data and embedded in a GPU-accelerated stochastic dynamic program that yields the minimum expected total cost across a 14-day shelf-life. Discrete-event simulation benchmarks the best (s,S) rule against this optimum. Results show a distinct operating threshold: When expected daily demand exceeds one unit per day, the (s,S) policy costs significantly lower, with the small residual driven by shortage and wastage penalties. The identified threshold supplies a practical screening rule: pool demand or redirect elective usage to increase expected demand past one unit and safely employ a (s,S) policy, while advanced optimization tools must remain reserved for facilities with persistently sparse demand.

Ìý

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