D15OPERATIONS

Scheduling and dispatching agent

For firms with field workers, technicians, or mobile service teams, optimizes the daily schedule: which jobs go to which workers, in which order, accounting for skills, travel time, customer windows, priority, and any constraints (e.g. some jobs need two people, some require specific equipment). Replans during the day as new jobs come in, jobs run long, or workers become unavailable. The pattern's value is replacing a dispatcher's manual scheduling — which works fine for a few jobs but breaks under volume — with continuous optimization. Different from D13 (which classifies and routes inbound work) because this pattern actively constructs and modifies schedules.

WHERE THIS FITS

BUSINESS SHAPES

B2B services

VOLUME THRESHOLD

Below 20 field jobs per day a month, the payback rarely earns the build. Patterns this shape reliably pay back at 100+.

FITS BEST

Field service (HVAC, plumbing, mobile health), home services, any human-to-location dispatch.

PAYBACK · 4-8 moBUILD · Medium-HighVALUE · $60k-$200kWHEN · >20 dispatches/day

FAILURE MODE TO DESIGN AROUND

Overbooking due to optimistic time estimates → reality-calibrate from history.

REQUIREMENTS · 7 REQUIRED

Requirements describe capabilities the pattern needs in your environment, not the vendors you must buy. Any system that fills a requirement satisfies it — that’s what makes the catalog portable across the long tail of SMB tooling.

job_intake
REQUIREDREADevent
Stream of jobs needing scheduling, both new and rescheduled.
DATA SHAPE
Job records with customer, location, type, estimated duration, required skills/equipment, time window, priority.
COMMONLY FILLED BY
- service ticket queue
- work order system
- scheduling intake form for new jobs
worker_roster_and_skills
REQUIREDREADrequest
Who's available to do work, what skills they have, where they're starting from, what their hours are.
DATA SHAPE
Per-worker: shift hours, skills, certifications, home base location, vehicle and equipment available.
COMMONLY FILLED BY
- workforce management system
- HR system with skill tags
- internal team roster maintained by dispatch
location_and_travel_data
REQUIREDREADrequest
Geographic data: where jobs are, where workers are, what travel time looks like. The hard constraint that most scheduling decisions hinge on.
DATA SHAPE
Coordinates for jobs and workers, travel time matrices, traffic-aware estimates if available.
COMMONLY FILLED BY
- mapping service for travel estimates
- GPS feed from worker vehicles
- address geocoding service
constraint_definitions
REQUIREDREADcorpus
The firm's rules and priorities encoded explicitly. Without this, the optimizer optimizes the wrong thing.
DATA SHAPE
Rules: hard constraints (skill required, time window), soft preferences (worker continuity for repeat customers, minimize overtime), priority weights.
COMMONLY FILLED BY
- configuration maintained by dispatch leadership
- small admin UI for constraint tuning
- documented business rules captured during build
schedule_output_destination
REQUIREDWRITEevent
Where the optimized schedule lands so workers and customers know what's happening.
DATA SHAPE
Per-worker daily schedule with jobs in order, expected arrival times, travel directions, customer-facing notifications.
COMMONLY FILLED BY
- mobile app each field worker uses
- dispatch dashboard showing the full schedule
- customer notification system with appointment confirmations
real_time_status_feedback
REQUIREDREADevent
Live feedback on job progress: starts, completions, delays, no-shows. Lets the pattern replan rather than running on yesterday's data.
DATA SHAPE
Status updates per job: started, in-progress, complete, blocked, with timestamps and any notes.
COMMONLY FILLED BY
- status updates from the field worker mobile app
- GPS-based completion detection
- customer-facing confirmation buttons
exception_decision_route
REQUIREDWRITEevent
Where decisions the pattern can't make alone get escalated: rescheduling that affects SLA, overtime authorization, customer accommodation requests.
DATA SHAPE
Exception with context, proposed options, financial or service impact, requested decision.
COMMONLY FILLED BY
- dispatcher review queue
- chat alert to the dispatch lead
- approval workflow inside the scheduling system

RUNTIME FLOW · 9 STEPS

01
Read all jobs needing scheduling for the period (typically next 1-7 days)
job_intake
02
Read worker availability, skills, and starting positions
worker_roster_and_skills
03
Build the travel-time matrix between all relevant locations
location_and_travel_data
04
Apply constraints and run the optimization to produce an initial schedule
constraint_definitions
05
Publish the schedule to workers and customer-facing notifications
schedule_output_destination
06
Through the day, ingest real-time status updates
real_time_status_feedback
07
When events trigger replanning need (job runs long, worker unavailable, urgent new job), recompute the affected portion of the schedule
DECISION Replan only the affected forward window; don't disrupt already-started jobs.
08
If replanning would require breaking SLA or business rules, route to exception decision
exception_decision_route
09
Publish updates with customer notifications for any changed appointments
schedule_output_destination

EMISSIONS · 3

Structured outputs this pattern produces. Other patterns and client systems can subscribe to them, which is how the catalog composes over time.

schedule_efficiency_signal
Utilization, drive time, overtime, SLA hit rate per worker and overall. The headline operational metrics.
CONSUMED BY
- operations dashboards
- manager monthly reviews
- capacity planning
constraint_violation_signal
When the optimizer couldn't satisfy all constraints, surfaced for review.
CONSUMED BY
- dispatch lead workflows
- policy refinement
- capacity planning
customer_appointment_data
Per-customer appointment history, useful for B5 churn-risk patterns and customer success.
CONSUMED BY
- customer success workflows
- service quality dashboards