Analyzing the Surgical Suite Environment by Focusing on Scheduling and Simulation Logic

by Zhanting Gao and Olivia Pewzer (HDR, Inc)

As presented at the 2017 Simio Users Group

HDR and Operations Design

For more than a century, HDR has partnered with clients to shape communities and push the boundaries of what’s possible. Our expertise spans 10,000 employees, in more than 225 locations around the world—and counting. Our engineering, architecture, environmental and construction services bring an impressive breadth of knowledge to every project. Our optimistic approach to finding innovative solutions defines our past and drives our future.

Operations Design is an integral component of HDR’s Consulting Group and contributes to our high-performance buildings and smart infrastructure. We work regularly and collaboratively on initiatives with professionals in predictive analytics, planning, research, strategic innovation and sustainability to eliminate inefficiencies in care delivery and to enhance the patient experience.

As a team of Lean Experts and Six Sigma Black Belts, we believe that containing costs is one of the most significant factors in helping to make healthcare more affordable, especially costs related to inefficiencies in how a building facilitates the work of its users, as well as, how it functions from an operations standpoint. Applying Lean Six Sigma concepts, along with a series of advanced modeling approaches, Operations Design investigates flows of resources and patients to determine the best operational scenarios and provide manageable and sustainable solutions.

Problems/Challenges and Why Simio Was Chosen

The Operations Design team was selected to complete an operational analysis in a surgical suite to be built in 2019. The OR suite is constituted of 3 main spaces: pre/post-op care space, OR suite and the post-anesthesia care unit (PACU). Currently, there are three proposed strategies to occupy this new perioperative space, and each strategy consists of an alternate surgery population with varying surgery volumes.

The client requested support in their decision-making process to determine which strategy best meets OR performance benchmarks, by taking into account ring growth and operational improvements. During the modeling and analysis, OR utilization and patient waiting times were selected as the main indicators of operational efficiency.

Given the number of scenarios to be analyzed and the multiple metrics of success, the Operations Design team decided to use discrete event simulation modeling for this client. Simio was identified as the most appropriate tool for this exercise because it offers:

  • Advanced discrete event simulation capabilities
  • Pre-set functions that require minimal modeling effort
  • Ability to quickly model various operational scenarios
  • Ability to import architectural drawings into a Simio facility window
  • Advanced 3D graphics to help easily obtain clients’ buy-in
  • Output results can be easily analyzed within other visualization tools

Approach

Given the nature of the analyzed space, this modeling exercise proved to be complex. Indeed, the operational goals and constraints placed upon the operating rooms by leadership really drive the arrival of patients, given the tight surgery schedule. Surgery scheduling is an entire analysis itself and has to be addressed and analyzed on its own. Therefore, we made the decision to build two different simulation models as described below:

  • A first model generates the scheduled time of the surgery under specific operational constraints: hours of operations, OR/specialty assignment and targeted OR utilization. This model generates an optimal sequence of surgery cases targeting 80 percent utilization in the OR suite. Without knowing the exact sequence of cases, it was not possible to consider only one model for the entire analysis.
  • Using the scheduled time of the surgery (from the first model), as well as, the time spent in pre-operative care, we determined the timing of the patient arrival used in a second model. The second model is a representation of the entire patient flow throughout the different spaces: pre- and post- operative rooms, operating rooms and PACU rooms.


Figure 1 Scheduling Model

After addressing the challenge of a complex scheduling process, the simulation specialists developed customized fixed classes in order to ease the process flow modeling effort. The customized fixed classes were designed to ensure that the entities involved have the ability to remain in a processing station while waiting for a specific event. This customization at the class level proved to be a key element of the simulation and eased the logic building, as well as the output result process. Finally, the queue storages played a crucial role in the model, as they allowed entity selection according to specific criteria. For instance, an OR suite is consistently driven and constrained by a different patient priority which requires complex patient selection logic. Queue storages proved to be the perfect elements to use for this exercise.


Figure 2 Customized Fixed Class

Such approach has some limitations and it was crucial to understand the trade-offs. The first model (generating the patient arrival) provides one possible combination of the surgery schedule over one year. As a next step, it became necessary to investigate how to automate the first model, in order to test different arrival patterns in a timely manner. Also, one of the many purposes of a discrete-event simulation effort is to test different routing probability. With the approach described above, it was not possible to vary the percentage of patients going to the pre-op care—currently set at 75 percent. If necessary, future investigations will focus on automating the creation of the data inputs to the second model, to allow more flexibility and variability in the analysis.  

A Targeted Result and Solution

Using Simio and the approach described previously, we were able to inform the decision-making process to provide space, enhanced the patient experience and resource-constraint indicators. Therefore, unexpected future space constraints (due to the high volume of patients designated to go the new space) were discovered and avoided. The clinical experts then understood the OR suite capacity, given their operational constraints, and ultimately made the decisions to:

  • Reduce by 25 percent the patient population intended to use the new OR suite
  • Merge the pre-op, post-op and PACU care spaces into one

In addition, the developed tool can be used to test different patient scheduling patterns, block schedule assignments, or even patient volumes, throughout a couple of future fiscal years. Moving forward, the intent is to use the Simio Enterprise version to investigate level-loaded surgery schedules.