Queueing and Applied Probability SIG

Welcome to the website of the 24th edition of the CanQueue workshop, which will be held on August 23-24, 2024 in Edmonton, Alberta, at the Lister Conference Centre of the University of Alberta.

• The CanQueue conference
• Organizing Committee
• Keynote Speaker
• Abstract Submission
• Registration
• Important Dates
• Accommodation Information
• About Edmonton

What is CanQueue?

CanQueue is one of the major queueing conferences in North America. The first workshop was organized by Dr. A. S. Alfa at the University of Manitoba in 1999, and the name CanQueue has been associated with these meetings since the 2000 edition in London, Ontario. The goal of the conference is to promote research and applications of queueing theory.

This annual conference provides an important platform for people, including leading Canadian and international queueing theorists, applied probabilists, scientists, researchers, engineers, executives, and students, to meet and share their new research findings, and to encourage collaboration on ongoing research initiatives.

Organizing Committee

• Saied Samiedaluie (samiedal@ualberta.ca) - University of Alberta
• Armann Ingolfsson (aingolfs@ualberta.ca ) - University of Alberta

Keynote Speaker

We are pleased to inform that Dr. Baris Ata (https://www.chicagobooth.edu/faculty/directory/a/baris-ata) from the University of Chicago’s Booth School of Business will be our keynote speaker this year.

Title: Control of High-Dimensional Queueing Systems in Heavy Traffic: A Computational Method Based on Deep Neural Networks Abstract: 

In this talk, I will first consider singular control problems that emerge as approximations of dynamic scheduling problems in heavy traffic. To solve them, I approximate them using drift control problems, extending the seminal work of Han et al. (Proceedings of the National Academy of Sciences, 2018, 8505-8510). I will then describe a general technique to implement the derived solutions of the drift control problems in the context of the original queueing networks, testing them against the best available benchmarks for dimensions up to d=30 or more. This talk is based on papers coauthored with Mike Harrison and Nian Si.

Bio: 

Baris Ata is the Sigmund E. Edelstone Distinguished Service Professor of Operations Management at The University of Chicago Booth School of Business. He takes a problem-driven approach to bridge theory and practice in operations management, focusing on dynamic decision-making in complex settings under uncertainty. On the theoretical side, Ata’s current research interests include solving high-dimensional stochastic control and learning problems. On the application side, he works on identifying suitable candidates for xenotransplantation, improving operations in the criminal justice system, and addressing logistical challenges in last- mile delivery in Africa. Ata’s research has earned several prestigious awards, including the Best Paper in Service Science Award from INFORMS (2009), the William Pierskalla Best Paper Award from INFORMS (2015), and the Wickham Skinner Best Paper Award from POMS (2019). He is also a recipient of the Manufacturing and Service Operations Management Young Scholar Prize from INFORMS (2015) and the Emory Williams (schoolwide) MBA Teaching Award at Chicago Booth (2021).

Abstract Submission

Presentations of ongoing or completed research in queueing theory, applied probability and related areas are invited.

We define “queueing theory” broadly, to include any type of analysis of service systems in which congestion could occur, using any type of methodology.

Please use this link to submit your abstract: https://forms.gle/1s6zD9TtEHMRuWgL9

Registration

You have two options for registration:

• Conference only: CA$150 + applicable taxes.
• Conference + Friday dinner: CA$225 + applicable taxes.

Please use this link to to register: https://forms.gle/RZVMzoo4Q2bnX2wE8

In order to help us with planning the event, we appreciate it if you could register as soon as you can.

Important Dates

• 12 August 2024: Registration deadline
• 12 August 2024: Abstract submission deadline

Accommodation Information

Lister Centre offers a limited number of hotel-style guestrooms. To check rates and availability, please see https://www.ualberta.ca/conference-services/accommodation/guest-rooms.h… .

Other options:

• Campus Tower Suite Hotel: Located close to the university campus. 14 minutes walk to the conference venue. https://www.campustower.com/
• Varscona Hotel on Whyte: Located on the historic district of shopping, dining, and entertainment of Edmonton. https://www.varscona.com/
• Downton hotels include JW Marriott, The Westin, Fairmont Hotel Macdonald, Courtyard Edmonton Downtown, etc. It takes about 20 minutes by LRT to get from downtown Edmonton to the university campus, or if you feel up to it, it takes a one-hour scenic walk, crossing over the North Saskatchewan river.

Edmonton, Alberta

Edmonton is a vibrant urban centre in the heart of the wilderness, the largest northernmost metropolis, and the capital of Alberta, Canada. It is known for its natural scenery, food, and history. Edmonton is also known as Canada's Festival City.

Here are some attractions you can enjoy during your visit to Edmonton:

• Fort Edmonton Park: Canada's largest living history museum.
• West Edmonton Mall: North America's largest shopping mall.
• Edmonton's river valley parks: The longest stretch of connected urban parkland in North America.
• Festivals: During the 2024 CanQueue's weekend, you can attend Edmonton Fringe Festival (the North America’s biggest and oldest fringe festival) and Edmonton PRIDE Festival.
• Royal Alberta Museum, Art Gallery of Alberta, Alberta Aviation Museum, Muttart Conservatory, and Alberta Legislature Building.
• Elk Island National Park: By driving only 50 minutes from downtown Edmonton, you will get the chance to visit the densest population of hoofed mammals in Canada (including bison, moose, deer and elk) and over 250 bird species.

Welcome to the website of the 23nd edition of the CanQueue workshop, which will be held on August 25-26, 2023 in Niagara-on-the-Lake, Ontario.

 

• Printed Program

• The CanQueue conference
• Organizing Committee
• Keynote Speaker

What is CanQueue?

CanQueue is one of the major queueing conferences in North America. The first workshop was organized by Dr. A. S. Alfa at the University of Manitoba in 1999, and the name CanQueue has been associated with these meetings since the 2000 edition in London, Ontario. The goal of the conference is to promote research and applications of queueing theory.

This annual conference provides an important platform for people, including leading Canadian and international queueing theorists, applied probabilists, scientists, researchers, engineers, executives, and students, to meet and share their new research findings, and to encourage collaboration on ongoing research initiatives.

Organizing Committee

• Amir Rastpour (amir.rastpour@uoit.ca) - Ontario Tech University
• Yiqiang Zhao (zhao@math.carleton.ca) - Carleton University

Keynote Speaker

We are pleased to inform that Dr. Cynthia Rudin (https://users.cs.duke.edu/~cynthia/) from Duke University will be our keynote speaker this year.

Title: How to Create Simple Risk Scores from Complex Models and Datasets Abstract: 

Consider predicting congestion at the intensive care unit (ICU) of a hospital, where overcrowding is life-threatening. While we would think this task is challenging due to heterogeneity of patients who have different diagnoses, complexities, and recovery times, as it turns out, simple linear machine learning models can predict congestion results of queueing models surprisingly well. This is true regardless of which queueing model is used for the ICU. In fact, simple models have been used across healthcare for decades. The vast majority of these models are scoring systems, which are sparse linear models with integer coefficients. Generally, such models are created without data, or are constructed by manual feature selection and rounding logistic regression coefficients, but these manual techniques sacrifice performance since humans are not naturally adept at highdimensional optimization. I will present the first practical algorithms for building optimized scoring systems from data. These methods have been used for several important applications to healthcare and criminal justice.

Bio: 

Dr. Cynthia Rudin is the Earl D. McLean, Jr. Professor of Computer Science and Engineering at Duke University. She directs the Interpretable Machine Learning Lab, and her goal is to design predictive models that people can understand. Her lab applies machine learning in many areas, such as healthcare, criminal justice, and energy reliability. She holds degrees from the University at Buffalo and Princeton. She is the recipient of the 2022 Squirrel AI Award for Artificial Intelligence for the Benefit of Humanity from the Association for the Advancement of Artificial Intelligence (the “Nobel Prize of AI”). She received a 2022 Guggenheim fellowship, and is a fellow of the American Statistical Association, the Institute of Mathematical Statistics, and the Association for the Advancement of Artificial Intelligence.