Opserv Two Large and Immediate

OpServ

Two large and immediate scheduling problems emerge from a cursory examination of the scheduling and patient data that would create problems with long wait times. First, though it is estimated that sixty percent of all patients arrive on a walk-in basis, more clinicians are devoted to appointment care throughout the day than to the handling of walk-in arrivals. That is, more patients are in demand of walk-in assistance, but more appointment clinician-hours are being offered. Switching this, or simply making all clinical staff more flexible -- setting appointments during hours that they will otherwise be seeing walk-in patients -- would go a long way towards solving the scheduling problems and the at time issues at the clinic. In addition, it does not appear that there has been any attempt to match clinician scheduling with patient patterns, with a great deal more variability in clinic staff than in patient visits and in fact more clinicians on staff during the last open hour of the clinic, when the fewest patients come in. In other words, a very basic rescheduling plan could be put into place without any in-depth analysis and still yield dramatic improvements. Brief statistical analysis and a bit of deeper logical thinking enhance these benefits still further, however, as the accompanying spreadsheet shows. Combining walk-in and appointment staff makes it clear that there are ample clinicians available to treat students in a more timely fashion if only they were better utilized, and in fact changes can be made to make everything more efficient.

b)

Cutting wait times to ten minutes or even less will be very simple, and determining a scheduling plan for such an outcome is not especially difficult. The exact parameters of the length of patient visits is not given, therefore this cannot be estimated with precision, however, with a given average of twenty minutes per clinician-patient visit and knowing that this follows an exponential distribution, it can be assumed that approaching 90% of visits will end within half an hour, and that a fair percentage will last a little bit less than twenty minutes, therefore a base rate of two patient visits per-hour per-clinician is a fair estimate. Calculating directly from the total patient visits at each clinic hour during the week, it can be seen that the maximum number of clinicians needed at any one time is eight, with only one patient waiting any significant length of time. The mean number of visits each day and the standard deviation can be used to asses a maximum number of expected patients; at 1.5 standard deviations higher than the mean (expected to contain 90% of all instances) the highest patient load is 22, which would require 11 clinicians. Assessment using the Poisson distribution gives slightly lower numbers, with 19 patients being the maximum load in ninety percent of cases, requiring 9 or 10 clinicians. The table provided calculates minimum staff needs based on real observations, resulting in no wait time for most patients.

c)