Start a PHOTOM dispatch scenario
FieldOpt is a PHOTOM maintenance scheduling optimizer for PV field operations, designed to improve dispatch decisions using urgency, skill matching, and travel-aware scheduling.
Choose a scenario
Typical weekday
A normal operating day across seven PV sites. A balanced PHOTOM maintenance backlog of electrical and mechanical faults, most technicians available.
Most jobs should get scheduled. Watch how high-priority faults land in the morning slot with travel-aware dispatch.
Short-handed day
Nine open PHOTOM faults across five sites, but only three technicians and limited slot availability. Some jobs will not fit.
Expect a few unassigned jobs. The optimizer should still prioritise high-impact electrical faults with lowest travel cost.
Post-storm surge
Eleven jobs across seven sites after last night's storm. Four technicians on shift, one available only in the afternoon.
Trade-off heavy: high-priority electrical work dominates the morning; low-priority tasks may slip. Travel-aware scoring keeps site clusters together.
What's happening?
PHOTOM maintenance backlog
Integration readiness for PHOTOM and COSMIC workflows
PHOTOM / COSMIC integration readinessMachine-readable ingestion of PHOTOM operational data.
- Maintenance jobs with priority, skill, duration, coordinates
- Technician availability, skills, and home base coordinates
- Monitoring-derived fault data (energy loss, affected capacity)
- JSON payloads today; CSV import ready to wire
Backend API + optimization engine.
- Transforms PHOTOM operational data into dispatch decisions
- Travel-aware scoring across technicians, slots, and sites
- Deterministic solve exposed over HTTP for pipelines and UIs
Structured outputs for downstream systems.
- Schedules with per-assignment technician, slot, and travel
- KPI summaries: travel, downtime, recoverable yield
- Scenario comparison for reporting and pilot review
- Ready to feed COSMIC dashboards and connected O&M tools
Each dispatch decision carries a rationale.
- Per-assignment reason string from the optimizer
- Human-readable — supports operator trust and adoption
- Traceable for reviewers and industrial stakeholders