4 Presentations

• 
  10:30 AM - 10:55 AM

  First or Second Doses First? Vaccine Allocation Under Limited Supply

  • Chaoyu Zhang, presenter, University of Toronto
  • Ming Hu, University of Toronto
  • Yun Zhou, McMaster University

  How to allocate limited two-dose vaccines, such as mRNA vaccines, between the first vs. second doses provoked a heated public debate during COVID-19 in January 2021. In this paper, we study the optimal vaccine allocation between the first vs. second doses with a constant stream of vaccine supply by formulating it as an optimal control problem under disease transmission to minimize the total number of infections over a planning horizon. Specifically, we extend the SIR model to incorporate the role of vaccines by adding two compartments, i.e., people who have received one dose and those who have received two doses. We demonstrate that the optimal vaccine allocation policy has a bang-bang structure: there exists a threshold on the one-dose vaccine efficacy that is higher than one-half of the two-dose vaccine efficacy, above (resp., below) which choosing the “First Doses First” (FDF) (resp., “Second Doses First” (SDF)) policy is optimal. Using COVID-19 vaccination data, we calculate thresholds for different countries in January 2021 to recommend to governments whether to use the FDF or SDF policy. Lastly, we demonstrate that our model can be extended to account for boosters by studying how to allocate limited vaccines between the second and booster shots.
• 
  10:55 AM - 11:20 AM

  Counterfactual Analysis in Complex Time Series Setting with Feedback Effects

  • Zhoupeng (Jack) Zhang, presenter, Rotman School of Management, University of Toronto
  • Opher Baron, Rotman School of Management, University of Toronto
  • Dmitry Krass, Rotman School of Management, University of Toronto
  • Tianshu Lu, Huawei
  • Marko Duic, Department of Emergency Medicine, Southlake Regional Health Centre

  Counterfactual analysis is an effective approach to evaluate policy impacts. Yet in the empirical Operations Management literature, standard errors for estimated impacts are oftentimes provided without clearly defined methodologies, rendering statistical significance of policy impacts ambiguous. In this research, we study dynamic counterfactual analysis using time series data with a focus on significance tests. In particular, we study settings with “feedback effects”, a fundamentally dynamic that features dependent and independent variables mutually reacting to historical variations over time. We first develop and estimate a simultaneous equations model (SEM) with general error covariance to capture the feedback effects. We then perform an iterative counterfactual analysis to estimate policy impacts in both the short- and long-term. Finally, and in particular, by exploiting the structure of SEM, we design a novel Bootstrap method to produce standard errors and construct 95% confidence intervals for the estimated impacts. We demonstrate by analyzing the impacts of a waiting time reduction project at the Southlake Regional Health Center near Toronto. We compare confidence intervals given by our Bootstrap approach with the ones derived from asymptotic approximation theory, and discuss the applications of our approach to other settings.
• 
  11:20 AM - 11:45 AM

  Data-driven patient scheduling for speech and language therapy considering cancellations and no-show

  • Sina Hoveida, presenter, Student
  • Mirhashemi Parmida, University of Waterloo
  • Hossein Abouee Mehrizi, University of Waterloo
  • Brendan Wylie-Toal, KidsAbility

  We present a data-driven approach for patient scheduling in a speech and language program to reduce patient wait times and improve program efficiency. Using historical data, we first develop prediction models for cancellations, no-shows, and the number of sessions required for each patient. We then propose a model for patient scheduling that takes into account cancelations and no-shows. Using data from a large healthcare provider, we demonstrate the significant impact of patient scheduling on system utilization.
• 
  11:45 AM - 12:10 PM

  Optimizing Operating Room Planning and Scheduling using Robust Optimization and Fuzzy Hyperheuristics: Addressing the Case Mix Problem

  • Justin Britt, University of Windsor
  • Ahmed Azab, University of Windsor
  • Fazle Baki, presenter, University of Windsor
  • Rifat Bin Hassan , University of Windsor

  In this presentation, the Case Mix Problem (CMP) from the strategic phase of operating room (OR) planning and scheduling is considered. This problem involves assigning the number of time blocks, number of recovery ward (RW) beds, and number of patients to surgeons in a way that both maximizes the social benefits of performing surgeries and optimizes the utilization of the sys- tem of ORs and RW beds. The CMP has several uncertainties associated with demand for surgeries, surgical durations, turnover times between surgeries, and patient lengths of stay. In the proposed integer programming model, robust optimization (RO) is used to account for uncertainties. The uncertainties and large numbers of variables and constraints in the RO model lead to tractabil- ity issues when using exact solution methods. Therefore, a hyperheuristic is developed and used to obtain solutions. The fuzzy late acceptance hyperheuris- tic (FLAHH) uses several different heuristic selection methods (simple random, choice functions, and reinforcement learning) and a late acceptance-based move acceptance method. A fuzzy logic approach is used to tune the length of the late acceptance list.