Construct a computer simulation model to replicate CT operations for decreasing delayed procedure starts.

Abstract

132 Background: Patients scheduled to undergo computed tomography (CT) should be treated expeditiously and not delayed owing to a lack of either CT scanner capacity or available staff. Delayed scanning affects both patients and staff in several ways. First, patients are unhappy that they have to wait. Also, delayed scanning makes patient late for their next appointments or other events, which affects the downstream departments’ capability to operate effectively and efficiently. In addition, radiologists and their staff have to commit additional time and resources to processing patients on time. Finally, variability in the placement of patients reduces the scanner’s operating efficiency. The aim of this initiative is to optimize the appointment template using simulation software to reduce the rate of delayed CT procedures by 25% or more by the end of 2014. Methods: To further understand the CT queuing process, we hired 2 graduate students to create a simulation model using the data collected from the operations study. The simulation study modeled patients’ experience from their arrival to discharge and the steps were: (a) performed elemental analysis for each process; (b) cceated value stream map; (c) created high-level simulation model and “mini model” using operational data. The simulation models were presented to department leaders, who approved them. The models clearly showed that the time patients spent on the CT scanner was the bottleneck. Results: Changes in the CT area that have impacted on-time starts and average wait time include: (a) new fast-track for no interview patients and (b) changes in staffing hours. Progress and improvement include (a.) On-time delays decreased by 18% and (b.) a verage wait decreased by 8 minutes (19%). Conclusions: Discrete event simulation accounts for the probabilities and uncertainties associated with the processes and helps create a visual model of the work area. This adds confidence to decision makers’ ability to make decisions that have high impact. Also, the models can be used to test changes in the processes and study the impact on other processes without making true operational changes that could potentially waste resources and time.