3 Presentations

• 
  09:00 AM - 09:30 AM

  Benefits of a Broader View: Patient Flow Modelling of Congested Hospitals

  • Carolyn Busby, presenter, ORCHID Analytics
  • Michael Carter, University of Toronto

  There are many examples in the literature of techniques aimed at optimizing bed capacity and allocation, reducing ED waiting time and maximizing operating room utilization; however, most draw narrow boundaries around a specific de-partment or service or tackle each problem in isolation. These approaches help diagnose department specific issues but miss the wider picture and fail to capture the upstream and downstream impacts. This is particularly important in congest-ed hospitals where operational issues in one department will likely have a ripple effect throughout the hospital. This inter-department dependence in congested hospitals is reflected in the surge policies used when congestion reaches a thresh-old that is deemed to require a whole hospital response. A hospital-wide generic simulation model, that factors in the effects of congestion, has been built to tack-le these problems. Insights on the importance of wide model boundaries and surge protocol modelling, gained through application of the model at three hospi-tals in Ontario, Canada, are shared. Example scenarios used to illustrate this in-clude early discharge planning, and surgical throughput improvement.

• 
  09:30 AM - 10:00 AM

  Reallocating operating room time: a Portuguese case

  • Mariana Oliveira, presenter, Instituto Superior Técnico, Universidade de Lisboa
  • Luísa Lubomirska, Instituto Superior Técnico, Universidade de Lisboa
  • Ines Marques, Universidade de Lisboa

  Health care providers face a continuous increase in the complexity of organizations mainly due to the increasing demand and to the development of new and expensive technologies. The operating room (OR) is a major challenge in the hospital and is crucial for the institution financial health. Moreover, the OR has a large impact in several units of the hospital and on the workforce of the immediate up- and downstream units. In the last decades, surgery demand has been increasing with restrictive resources, forcing ORs to be more efficiently and effectively managed. This work is developed under a partnership with a Portuguese public hospital and aims to achieve a major social impact, which is increasing surgical access and thus reducing the patients waiting lists. Given the hospital restrictions in terms of space and staff, this work focuses on the reallocation of the available OR time among the surgical services, proposing new master surgical schedules - aggregate production planning consisting of timetables with specific timeslots assigned to each specialty. The main objective is to match demand and the existing capacity while maximizing OR efficiency. This work proposes a mathematical programming model, with three objectives: to maximize the allocated slots weighted by aggregated staff preferences; to match supply and demand; and to level the workload of up- and downstream units. A comparison of the actual allocation of slots with the one suggested by this approach is performed. Results show that the workforce is one of the major bottlenecks, suggesting a new distribution of the workforce among the specialties.

• 
  10:00 AM - 10:30 AM

  Stochastic surgery scheduling under a balanced ratio of in- and outpatients and ward uncertainty

  • Asgeir Orn Sigurpalsson, presenter, University of Iceland
  • Thomas Philip Runarsson, University of Iceland
  • Rognvaldur J. Saemundsson, University of Iceland

  Hospitals worldwide are faced with the pressure of an aging population and increased costs while needing to deal with the lack of nurses. The response has been to increase the overall utilization levels of existing resources. In this work we address the surgery scheduling problem and the problem of the unbalanced flow of patients and thus the risk of last minute cancellations. A novel approach to create a master surgery schedule for the operators which maximize throughput but simultaneously balances the flow to the wards will be presented using a two step approach.In the first step, a Monte Carlo simulation is used to create all feasible combinations of operating room days from each operator’s waiting list. As many combinations are created we limit the search space such that each combination i)forces a correct ratio between in- and outpatients for each operator and ii)forces the probability of running into overtime is bound. A mixed integer programming model is then proposed to schedule the combinations to room and days while bounding the probability of going over the available ward beds downstream. The approach is inspired by real challenges and tested on real-life hospital data.