Your reports were full of errors, assumptions and manipulated data. Your first error: Our lunches are only 45 minutes paid. Not one hour as you assumed and erred. For example: We come in 15 mins early for lineup making our day 8hrs 15mins. Take away one hour for lunch and that equals 7hrs 15mins. We get paid for 8hrs. Subtract 7hrs 15mins from 8hrs and you get 45mins for lunch.

Your basic math is wrong and you are making assumptions because we have one hour for lunch. If you aren’t able to calculate simple numbers, how well did you perform with the rest of the more complex calculations?

Also, our stats are manipulated because our 9-1-1 system is so outdated. I just attended the CALNENA event last month and interviewed phones vendors who understand the large limitations of our old systems. Inaccurate stats was the tip of the iceberg. I didn’t see anything in your report mentioning this crucial factor.

As a 3rd party who doesn’t seem familiar with our agency, you should have asked a lot more questions and talked to more people before submitting your report.

Matthew

Dispatcher Analysis of Office of the Controller – City Services Auditor City & County of San Francisco Report dated January 10th, 2014

Summary:
Of the 68 pages that make up the primary portion of the report issued by the Controllers Office, 32 of those pages have one or more facts, figures or data sets that are in error, or approximately 47% of the report. (Appendix F is primarily an overview of the day to day scheduling operations, and was not included in the above figure due to the amount it repeats the same information while describing the complexity of scheduling.)
The first page of the report states DEC is overstaffed even though one of the recommendations would be to have less than 1 dispatcher (0.3) assigned to call taking duties during certain times of the day for a city of over 825,000 people.
Several of the recommendations the report makes are similar to an agenda Management has been intending to implement during Labor Management and Meet and Confer sessions for years. Such as, among others:
Reduce staffing numbers
Reduce wrap time
Consider “alternatives” to lunch schedules
The report itself mentions in several places that the information used has been fabricated, that complete information is not available, and some data sets were thrown out completely. Half of the “peer agencies” chosen actually function very differently than San Francisco does and are a poor comparison. The method other departments use to calculate workload could be drastically different then the one used by DEM, which may explain such obvious discrepancies. Each community, each police force and each fire department are going to function uniquely due to local custom and demographics, so it of little surprise that finding a true “peer” to DEM would be difficult, if not impossible.
Given the potential that a report of this nature, could be used to influence city policy and political decision making, it is important that it reflect accurate information. The main bullet point of the report, to reduce staff, is highlighted on the cover page and has potentially serious public safety ramifications. The rest of the report could attempt to be used to have bearing on other important factors such as work place conditions, stress levels, and rights provided by law. There is concern that a “lay-person” not familiar with the dispatch environment could construe the bullet points without delving deeper into analysis of the information provided.

Assessing Minimum Staffing Levels:
The report attempts to assess minimum staffing levels based upon workload and the amount of time dispatchers are available for work. Each of the figures used are based on false information, as follows.

Annual Net Available Work Hours:
Page 9. The yearly estimated minimum hours for POST CPT is listed as 4, when it is actually 12. This is an additional 4% that a dispatcher is unavailable due to mandatory training.
Page 11 states the report calculated the amount of hours available to work based only on 8 hour shifts. That would add in an additional lunch break to time not available per week, or 52 hours a year for each 10 hour shift.
The report also assumes dispatchers are provided a 60 minute lunch period, when in fact it is 45 minutes. The additional 15 minutes adds 65 hours per year to time available for work.
The report does not take into account lunches and breaks for dispatchers on vacation. 15 hours for lunch and 7.5 hours for breaks get added in to time available for work.
Recalculating this information means the 392 hours listed for Breaks and Lunches becomes 252 hours and drops from 45% time not available to work to 35%. Total Annual Unavailable Hours goes from 873.5 to 707 and Annual Net Available Work Hours changes from 1212.1 to 1378.6, a difference of 166.6. Total percentage of time Dispatchers are available to work changes from 58% to 66% per year.
The report states that its own figures are not out of line with other San Francisco public safety departments. With accurate figures, DEM is then actually ahead of the rest when it comes to being available for work.

Assessing Workload:
Per page 5 of the report, calls are received through the Symposium System, however this report calculates workload based on calls logged in CAD. Not all calls received are logged into CAD even though a great deal of time may be spent helping that person who called in; for instance a citizen calling in on the non-emergency line with a question that does not require a field unit being dispatched on.
Several calls are received where the Call Taker uses the “on-view” command, such as providing a case number, and this report discards any call made as an “on-view”.
Most incidents from every priority range receive multiple calls. Such as several witnesses calling at the same time for a vehicle accident, or several people calling for several hours regarding a noise complaint. Each one of these calls is counted as only one call in this report. The report claims to use a “multiplier”, however that does not include “C” priority calls or the amount of time processing duplicate calls.
The report attempts to figure the amount of time a call taker may spend on each call by using time stamps in CAD. It measures from Initiate (time call received) to the Entry (time call sent up for dispatch) time stamps, however it does not figure the time spent collecting further information after the Entry period. For most calls, the time between Initiate and Entry will be very brief because the majority of the time is spent sending supplemental information after the fact. For instance, to get the call up for dispatch without delay, the first Entry may be “Someone just got shot”, followed by several minutes worth of processing time afterwards collecting scene safety information and victim condition while working with a distraught caller. Many callers hang up without providing any additional information, necessitating call backs which take up more processing time.

Staffing Model Recommendations:
Even though this report has no problem making recommendations, it acknowledges that it is unable to capture much of the information and that estimates have been used.
“In practice, DEM may not be overstaffed, and, in fact could be understaffed…”
From the hours of 0400 till 0700 the report recommends less than 1 dispatcher be required to take calls.
Example: Hour 5, 1.32 Positions Needed, Staffing Requirement 2.3. If 1 person is covering Drops, 1 person is on break, that leaves 0.3 of a person to cover phone calls.

Meal periods and relating to “peer agencies”:
The report assumes we have a 60 minute paid lunch break, when our lunch break is 45 minutes. The report suggests dispatchers be cut to a 30 minute lunch break. Due to the compensation for line-up, the minimum lunch could be reduced to is 45 minutes total. The report suggests that the department could increase dispatcher availability from doing this by 130 hours per year or 11%. Taking into account our lunch is already only 45 minutes, the maximum savings to the Department per year is only 4.7%, not 11%.
The report goes onto suggest that DEM “peer agencies” do not offer a paid lunch, when in fact only 2 of them do not offer a paid lunch.
The report also suggests limiting the amount of shift start times from the current 11 to follow suit with other agencies. More start times are more likely to make shift scheduling dynamic and adaptable to changing needs for the department.
The report also suggests DEM follow “peer agencies” by utilizing part-time staff. Only one of the “peer agencies” uses part-time staff.
The report suggests a change to an electronic based scheduling system. However, only one other “peer agency” uses this method, most using a paper-based system.
In collecting data from the “Peer Agencies” the report states that the questioning was changed several days into the process. Further, the Controllers Office “recoded” some “inconsistent responses”. Agencies contacted were only asked to estimate their call volume and the figures were not verified. Leaving no way to correlate how the data was collected with the way DEM’s data is collected, making for an inaccurate comparison.
Out of the 118 agencies the group was left with, 6 were chosen as “Peer Agencies”. Out of that group 2 only take emergency calls and 1 transfers all fire and medical calls to another agency, making for a poor comparison to the way DEM functions. Washington DC is the only agency similar to San Francisco by training staff in multiple disciplines.

In Conclusion:
The San Francisco Controllers Office is accountable to the citizens of San Francisco for providing reports with accurate information. A report produced where 47% of the pages within contain 1 or more errors can only be described as fraud. The Controllers Office is also responsible for managing the San Francisco Whistleblower program for reporting fraud throughout the city.