(Erin Hopper, Abby Sly, Meagan Simmons, 2001)
The following report discusses a simulation study conducted on the block scheduling techniques used in Dr. Roy Green's pediatric dental practice in Fort Myers, Florida. The purpose of this study was to analyze the current block scheduling procedure in order to determine the efficiency of Dr. Green's scheduling system in terms of chair idle time, patient waiting time, and server idle time. The study began by researching various papers on scheduling techniques for the purpose of applying appropriate methodology to our own study. This included, among other things, the justification of the use of simulation as the most accurate and efficient way to study scheduling problems and the approbation of the use of block scheduling in pediatric dental offices.
After the research was complete, a collection of historical data was necessary to create the experimental environment and the parameters under which the simulation would be run. By gaining a thorough understanding of the current block scheduling system, along with a detailed description of dental procedures and staff allocation, this knowledge allowed for the creation of several assumptions useful in both clarifying and simplifying the model. This insight also permitted the application of variability and appropriate statistical distributions to the simulation that mimicked the existing dental office environment.
With the use of the ProModel simulation software package, the simulation was developed to model Dr. Green's real world scheduling practices. In this basic queue system, patients arrive, are greeted by the receptionist, and escorted to dental chairs where they have the appropriate and predetermined procedures performed by either a dentist or hygienist. ProModel then gathers statistics and reports on the model's distinguishing characteristics, including chair idle time, patient wait time, and server idle time. Dr. Green's block schedule is different for each day of the week, thus five versions of the simulation were necessary to portray the model accurately. In addition, the simulation for each day of the week was run five times in order to obtain results representing a full month's worth of scheduling.