pilipiliÂþ»­

 

Graduate Student Research Seminar Day ‑ Nov 19, 2020

You are cordially invited to theÌýGraduate Student Research SeminarÌýof theÌýDepartment of Industrial Engineering

Date:ÌýThursday, November 19, 2020
Time:Ìý2:00 - 4:00 PM
Venue: Online Event

Ìý

Abstracts:

Optimal transport scenario for access to endovascular therapy with consideration of patient outcomes and costÌý
Ashley Wheaton, MASc. Student

Stroke is a devastating disease for which time to treatment directly impacts patient outcomes. Ischemic strokes make up 75% of all strokes and are caused by an occlusion in a vessel of the brain that restricts blood flow to subsequent areas of the brain. There are two treatments for ischemic stroke: medical treatment (tPA) and endovascular therapy (EVT). EVT can only be used to treat ischemic strokes caused by an occlusion in a large vessel of the brain. EVT is only offered at hospitals with sufficient resources; therefore, there is uncertainty about the best transport destination for patients that may be eligible for EVT. A Drip and Ship protocol for these patients means transporting the patient to the nearest hospital for tPA then transferring them to an EVT enabled facility. A Mothership protocol means bypassing the nearest hospital and transporting the patient directly to the nearest EVT enabled hospital for tPA and EVT. A mathematical model was developed to assess which transport protocol is best considering both patient outcomes and transportation costs. The results of this model show that in some scenarios these metrics converge to a decision on which transportation protocol is best, and in other scenarios these metrics are divergent in their decision. In divergent scenarios patient outcomes can be improved at a greater cost to the health system

Ìý

Optimization of forecasting and scheduling practices in poultry procurement
Kenneth Dolson, MASc. Student

Eden Valley Poultry is one of Atlantic Canada’s primary suppliers of chicken and turkey products, purchasing large quantities of poultry from over 100 barns across Nova Scotia and Prince Edward Island. Coordinating flock procurement from these barns is complex, demanding a balance of meeting customer requirements and farmer production goals while ensuring the production facility’s constraints are not violated and a collection route will be feasible. A time series model for forecasting average flock weight is developed and used to inform a Chebyshev goal program which schedules flock pickups. This process is encapsulated in a procurement management tool designed for the client. Considerations are then made for how stochastic optimization could improve this model by using chance constraints to account for uncertainty in the predictions from the time series model.

Ìý

Deep learning-based stacking method and generative adversarial networks for human activity recognition based on ambient sensors
Qixuan Zhao, MASc. Student

Smart home for healthcare services has acquired more attention since the increasing development of the Internet of Things and the population ageing over the world. Human activity recognition (HAR) is one of the concerns of the smart home. Ambient sensors based HAR is one promising direction. This research proposes a deep learning-based stacking method for HAR based on ambient sensors. We first generate base models of convolutional neural networks (CNNs) and long short-term memory (LSTM) with different architectures, training data, and sliding window sizes. These base models are further integrated by a LSTM model to make final predictions. Furthermore, we propose a generative adversarial network to generate synthetic data as supplementary training data to tackle the problem of insufficient data. These two methods are used together on six real-world datasets. Results show that our proposed methodology statistically outperforms other works.

Ìý

Optimal pricing and production strategies for new and remanufactured products under a non-renewing free replacement warranty
Zhuojun Liu, PhD. Student

Stroke is a devastating disease for which time to treatment directly impacts patient outcomes. Ischemic strokes make up 75% of all strokes and are caused by an occlusion in a vessel of the brain that restricts blood flow to subsequent areas of the brain. There are two treatments for ischemic stroke: medical treatment (tPA) and endovascular therapy (EVT). EVT can only be used to treat ischemic strokes caused by an occlusion in a large vessel of the brain. EVT is only offered at hospitals with sufficient resources; therefore, there is uncertainty about the best transport destination for patients that may be eligible for EVT. A Drip and Ship protocol for these patients means transporting the patient to the nearest hospital for tPA then transferring them to an EVT enabled facility. A Mothership protocol means bypassing the nearest hospital and transporting the patient directly to the nearest EVT enabled hospital for tPA and EVT. A mathematical model was developed to assess which transport protocol is best considering both patient outcomes and transportation costs. The results of this model show that in some scenarios these metrics converge to a decision on which transportation protocol is best, and in other scenarios these metrics are divergent in their decision. In divergent scenarios patient outcomes can be improved at a greater cost to the health system.

Contact Person:
Prof. Dr. Floris Goerlandt
email: floris.goerlandt@dal.ca