MICHAELMICHAEL · Patrol workload and beat design
Officers per thousand residents was discredited decades ago.
It cannot see call volume, service time, or what your officers actually do with a shift. Most staffing studies still run on it. MICHAEL does patrol workload analysis from your own CAD export, and answers the two questions a staffing study exists to answer: how many officers the demand supports, and where the beat lines should go.
Your request competes with potholes.
A city manager is not weighing your staffing study against another staffing study. They are weighing it against a fire engine, a road resurfacing, and the position the parks department also wants. They do not care how the number was computed. They care whether they can explain it to a voter in one sentence.
- 01Most staffing cases do not fail on the analysis. They fail in the room. The work is sound, the slides are dense, and nobody in the audience can tell which numbers were measured and which were chosen. A council that cannot see the difference defaults to no.
- 02The ratio survives because it fits in a sentence. "We are at 1.8 officers per thousand and the state average is 2.2." No math, no assumptions, nothing to attack. It is also wrong, and most of the people in the room half know it is wrong, which is a strange place to be arguing from.
- 03A demand study only wins that room if the assumptions are visible. When someone asks why the relief factor is 1.6, "that is what the consultant used" ends the conversation badly. Every judgment call in this one sits on the panel as a number you can move while they watch.
How many officers the demand actually supports.
That ratio cannot see call volume, service time, or anything your officers actually do with a shift. Most staffing studies still run on it. This one answers the question from your CAD export instead.
- 01Dispatched and self-initiated work are shown apart, and never added together. Dispatched work arrives whether or not anyone is free. Patrol checks and traffic stops are what officers do with the time that is left over. Add the two and you overstate how busy the agency is while hiding the slack that proves you have room. On the first real extract we ran, patrol checks and traffic stops were 42% of all rows.
- 02One row per responding officer. A call held by three units for forty minutes is two officer-hours, not forty minutes.
- 03It adapts to your export, not the other way round. It reads the roles you assign to columns rather than their names, so another agency's export needs a re-designation and not a new version of the software. It uses the call source your CAD recorded; where your CAD recorded none, it guesses from the call type and you correct it in a click. Where the two disagree, the screen says so and names the types.
- 04Service times as median and 90th percentile, never an average. The distribution is right-skewed and the long tail is what eats a shift.
- 05Demand by hour and day of week, averaged over each weekday's own occurrences. A date range holding five Fridays and four Mondays would otherwise make Friday look 25% busier.
- 06Rejected rows counted by reason and shown. Negative durations, zeroes, anything past a ceiling you set. A workload total that quietly drops rows is wrong in the direction nobody checks.
- 07Capacity read from the same extract, no roster file needed. A unit that appears on a date worked that watch, so scheduled hours come from the export itself and the watch spans are read from your data rather than typed in. Saturation is committed time against net available, and out-of-service hours come off the denominator, where they belong.
Two bars, never one. Self-initiated was 42% of all rows on the first real extract, and folding it in would have counted the agency's spare capacity as capacity consumed.
The skew is why the median is the honest number: a mean gets dragged right by the long calls and reports a shift nobody worked. A sample CAD extract ships with the app, so you can see all of this run before you export anything of your own.
How many officers, and where the lines go.
The two questions a workload study exists to answer. Both of them get taken to a council most years with nothing underneath them.
- 01The staffing answer, with its assumptions on screen. How many positions the demand supports, which hours cross the Rule of 60 line, and which beats run hot. Every judgment call behind it (relief factor, threshold, staffing level) sits on the panel as a number you can move, not in a footnote. On the sample it finds one beat at 87% against another at 17% with positions in hand force-wide, and names that a reallocation finding, not a hiring one.
- 02Beat balancing that moves boundaries, not officers. Reporting districts are the atoms; reassign them between beats and every figure recomputes. It balances on citizen-generated committed hours, never a call count, and there is deliberately no optimizer: you know which streets are barriers and which neighborhoods must not be split. The tool does the arithmetic, which is the part people get wrong by hand.
A 70-point spread across one force. That is a reallocation finding, not a hiring one, and the difference is the whole argument in a budget meeting.
The analyst moved the districts. MICHAEL did the arithmetic, which is the part people get wrong by hand. There is deliberately no optimizer.
Same study, two presentations. One of them gets funded.
This is the part nobody trains for. The analysis is finished, it is correct, and how you say it decides whether it survives.
- 01The version that dies. "We computed saturation against net available hours, applied a 1.6 relief factor and a 60 percent threshold, and excluded rows with negative durations." Every word of that is true. It describes a method to people who are being asked to approve a decision.
- 02The version that gets funded. "Beat 3 is committed 87 percent of the time. Beat 2 is committed 17 percent. Same force, same shift, same night. We are not asking you to hire anyone. We are asking to move four reporting districts." That is a decision a council member can repeat at a town hall, and defend.
- 03The method still has to be behind it. The second version only holds up because the first one is true and sitting one click away. When the question comes, and it comes from the one member who reads everything, you open the panel and move the threshold while they watch.
What you will be asked, and what the honest answer is.
Four of these come up almost every time.
- 01"Why not use officers per thousand? Everyone else does." Because two agencies with the same ratio can have opposite workloads. The ratio counts residents. It cannot see how many calls arrive, how long they hold a unit, or what time of night they land. Your council is comparing you to a city whose calls you have never seen.
- 02"Our CAD does not flag self-initiated work." Plenty do not. It uses the call source your CAD recorded; where your CAD recorded none, it infers from the call type and you correct it in a click. Where the recorded source and the inferred one disagree, the screen says so and names the types rather than quietly picking one.
- 03"So how many people do we hire?" Sometimes none, and that answer is worth more than it sounds. On the sample extract the spread runs from one beat at 87 percent to another at 17 percent with the positions already in hand. That is a reallocation finding. Walking into a budget meeting and not asking for money buys you credibility you can spend later.
- 04"Can we check your math?" Yes, and that is the point of showing the rejected rows. Negative durations, zeroes and anything past a ceiling you set are counted by reason and displayed. A workload total that quietly drops rows is wrong in the direction nobody thinks to check.
A study you run once is a document. Run it every year and it is a baseline.
The first one is the expensive one, because you are arguing the method and the finding at the same time. After that you are only arguing the finding.
- 01Pull the same export and report the same numbers. Committed percentage by beat, median and 90th percentile service time, the dispatched and self-initiated split. Not a new set of metrics each year, chosen to flatter whichever direction things moved. The same four, every time, even in the years they are unflattering.
- 02Show whether the last decision worked. If you moved four reporting districts, the honest follow-up is whether the spread actually closed. Most agencies never go back and check, which is exactly why the next request gets treated as a fresh argument rather than a track record.
- 03A council that has seen the same chart three years running stops auditing the method. They start asking what the number means. That is a much better meeting, and it is the whole return on doing this properly the first time.
Try it on your own data.
Fourteen days, every feature, on one machine. Tell us who you are and we will email you your own download link. Nothing to activate. Your data stays on your machine and stays yours when the trial ends. No newsletter, no drip sequence.
Built for the people who do the work Sworn agencies, crime analysis units, public safety partners, and corporate investigation teams. Workstations are $2,500 a year, everything included. Academia and nonprofits: ask about license pricing.
Or email info@sworn.ai. A person answers.