A facility manager running one site can walk over and look at the generator. A facility manager running ten sites across three states is stuck calling site engineers one by one, asking "is everything okay?" and hoping the answer is complete. Fleet monitoring exists to close that gap — not by adding more people making more calls, but by putting every site's status, on the same timeline and the same threshold logic, in one place.
The Multi-Site Visibility Problem
The failure mode isn't dramatic — it's slow erosion of trust in status updates. A site engineer says the DG is fine because nothing's actively alarming, not because they checked the load trend. Multiply that across ten sites and "everything's fine" stops meaning anything specific. The facility manager finds out about a real problem only when it's already an emergency, days or weeks after the underlying issue started — and by then, reconstructing what happened means calling the same site engineer and asking them to remember.
Site-Level Rollup vs Device-Level Drill-Down
Fleet monitoring works the same way single-site monitoring does, just with an extra layer on top: a fleet view shows which sites are healthy and which need attention, and clicking into any site shows the same device-level detail — the same eight-parameter readouts for a DG or transformer — you'd get on a single-site dashboard. Nobody has to choose between "see everything" and "see enough detail to act"; both views pull from the same live data, queried at different scopes.
Why a Fleet-Wide Threshold Doesn't Work
A generator in Chennai and a generator in Shimla don't run at the same ambient temperature, so their coolant-temperature baselines are genuinely different — a reading that's a clear warning sign in Shimla can be an entirely normal Tuesday afternoon in Chennai. A transformer's oil temperature baseline shifts with ambient conditions the same way. Apply one generic threshold across the whole fleet and you get one of two failure modes: it's tuned loose enough to avoid false alarms at the hottest site, which means it misses real problems everywhere else, or it's tuned tight enough to catch real problems at the coolest site, which means every other site drowns in false alarms. Per-site threshold tuning against each site's own operating baseline is what makes fleet-wide alerts trustworthy rather than something teams learn to ignore.
Who Sees What: Role-Based Access Across Sites
An HQ operations head needs the whole fleet at a glance. A regional manager needs their five sites in more detail. A site engineer needs their one site in full depth, in real time. Role-based access — superadmin, admin, and site-scoped monitor roles — means all three are querying the same underlying dataset, filtered to what's relevant to their role, instead of three people maintaining three separate spreadsheets that inevitably drift out of sync with each other.
Connectivity Gaps Don't Mean Data Gaps
Multi-site fleets are exactly where connectivity is least reliable — a generator yard or a remote cold-storage dock often has no dependable Wi-Fi at all. That's why monitoring is cellular-first by default. And when even cellular drops, the gateway buffers up to 30 days of readings locally and syncs the full backlog the moment connectivity returns — a short outage shows up as a delayed batch in the history, not a hole in it. For a fleet manager relying on that data to catch a slow-developing problem across a dozen sites, that continuity is the difference between a trend line you can trust and one with silent gaps exactly where something might have been happening.
A Worked Scenario: Catching a Problem Before the Site Engineer Notices
Say a fleet of ten sites includes one transformer whose THD (total harmonic distortion) has been creeping up over three weeks — a sign of degrading power quality that a site engineer checking voltage once a day wouldn't necessarily flag, since voltage itself can look normal while THD climbs underneath it. On a fleet dashboard with per-site baselines, that site shows up as a deviation from its own three-month average well before THD reaches a level that damages connected equipment — giving the regional manager a scheduled inspection to arrange, instead of a call three weeks later explaining why several machines on that transformer's circuit are showing premature wear.
Frequently Asked Questions
There's no hard site cap built into the dashboard — the same rollup-then-drill-down structure and role-based filtering applies whether it's covering one site or fifty, since each site's data is scoped independently and aggregated at query time rather than requiring a different architecture at scale.
Yes — role-based access (superadmin, admin, and site-level monitor roles) means a regional manager sees the sites assigned to them, a site engineer sees their own site in full detail, and HQ sees the full fleet rollup, all filtered from the same underlying data rather than generated as separate exports that can drift out of sync.
Monitoring is cellular-first (3G/4G/5G) by default specifically because facility Wi-Fi is inconsistent across sites — a generator yard or transformer enclosure often has no reliable Wi-Fi at all, so cellular avoids that dependency entirely rather than working around it.
The gateway at that site buffers readings locally — up to 30 days' worth — and syncs the full backlog the moment connectivity returns, so a temporary outage doesn't create a permanent gap in that site's history. It shows up as a delayed batch of readings, not a missing week.
Each site's thresholds are set against that site's own operating history rather than a single fleet-wide default — a transformer running consistently at 45°C oil temperature in Shimla and one running at 58°C in Chennai can both be perfectly healthy for their conditions, so alerts fire on deviation from each site's own baseline, not a number picked once and applied everywhere.