The Fuel Balance Problem
Every multi-site fuel operator runs a reconciliation at some interval — daily, weekly, or monthly. The reconciliation asks: how much fuel did we receive in deliveries, and how much did we sell? The two numbers should match, minus a small allowance for temperature-driven density variation, pipeline heel, and hose drip.
They rarely match cleanly. The variance — the difference between received and sold — is the fuel balance problem. At a well-run site, it might be 0.2–0.3% of total volume. At a site with a meter calibration issue, a slow subsurface leak, or fuel theft, it can exceed 1–2%.
The problem is that manual reconciliation catches the variance at the end of the measurement period — after 30 days, after 7 days, after whatever the operator’s reconciliation cadence is. By that point, the variance is a history, not an active condition. Root cause investigation requires reconstructing what happened across a period where daily measurements weren’t taken.
IoT-connected tank monitoring with continuous level measurement changes the reconciliation from periodic to continuous. Every delivery is recorded as it happens. Every sale is matched against level movement. Variances appear daily, in the days they occur, when investigating them is still practical.
The Four Sources of Fuel Variance
Understanding where the gap comes from determines which monitoring capabilities address it.
Source 1: Measurement Error in ATG Systems
Automatic Tank Gauges (ATGs) are the standard for compliance monitoring at fuel retail sites. They measure fuel level via a float probe and compute volume from the tank’s geometry table. ATGs are accurate when properly calibrated — but calibration drifts over time, probes can develop sticking or binding, and geometry tables may not account for tank settling or deformation over years.
An ATG reading “fuel on hand: 6,240 gallons” may be accurate to ±30 gallons or ±150 gallons depending on calibration and probe condition. Over a month of 80,000-gallon throughput, 150-gallon ATG error represents 0.19% variance — invisible in isolation, significant in aggregate.
VX-Olympus integration with ATG systems (Veeder-Root TLS, Franklin Fueling, Isco, and others) captures ATG readings at configurable intervals, tracks calibration error over time by comparing multiple ATG readings against delivery manifests, and flags when ATG accuracy appears to be drifting.
Source 2: Delivery Variance
A fuel delivery is measured at the truck’s meters — a certified measurement point. The gallons hitting the tank should equal the gallons leaving the truck. In practice, small variances occur:
- Pump cutoff timing (fuel in the hose when the delivery ends)
- Temperature differential between truck-measured and tank-measured fuel (fuel density varies with temperature; a gallon at 60°F is a slightly larger volume than a gallon at 85°F)
- Delivery to wrong tank (rare but documented — wrong grade fuel delivered to a tank)
VX-Olympus records the tank level at delivery start, monitors the level rise during delivery, and records the level after delivery stabilization. The computed delivery volume (tank level change × tank geometry) is compared to the delivery manifest volume. Consistent variances of the same direction suggest systematic measurement error; inconsistent variances suggest random events requiring investigation.
Source 3: Slow Subsurface Leaks
EPA regulations require ATG systems to perform automatic leak detection — both precision tank testing (when the site is quiet) and continuous leak detection during operating hours. The EPA’s minimum leak detection thresholds are 0.1–0.2 gallons per hour for precision tests.
A leak at 0.08 gallons per hour is below the EPA reporting threshold but still loses 57 gallons per month — roughly $200 at retail prices. Over a year, that’s $2,400 at one site. Across a 20-site portfolio, $48,000 annually in a leak that was never reported to regulators because it stayed below threshold.
VX-Olympus’s continuous level monitoring analyzes the level-vs-time trend during quiet periods (overnight, between deliveries) to detect anomalous level drops. Level drops that aren’t consistent with evaporation models or ATG calibration variance trigger an investigation flag.
Source 4: Fuel Theft and Meter Fraud
Fuel theft at multi-site operations occurs in several forms:
- Employee dispensing fuel into personal vehicles or external containers off-hours
- Meter manipulation (pumping against a stuck meter, manipulating pump firmware on older equipment)
- Delivery shortfall (driver delivering less than manifested and diverting the balance)
VX-Olympus monitors for unusual dispensing patterns: fuel leaving the tank (level drop) when no dispenser sales are recorded against that tank. A level drop overnight, when the site is closed, with no delivery recorded, is an investigation flag. A level drop larger than the dispenser sales records can account for is a dispenser or meter anomaly flag.
VX-Olympus Configuration for Fuel Operations
ATG Integration
VX-Olympus integrates with ATG systems via their serial or TCP/IP output protocols. Veeder-Root TLS-350, TLS-450, and Franklin Fueling i-Site systems are common integration targets. The integration reads:
- Current fuel level (inches or gallons)
- Water level in tank
- ATG alarm status
- Product temperature
- Probe status
Data is read at 15-minute intervals during normal operation, more frequently during deliveries (triggered by detecting rapid level rise). The ATG’s current readings are supplemented with VX-Olympus’s independent analysis — delivery volume calculation, leak rate analysis, variance tracking.
Multi-Site Dashboard
A fuel operations manager overseeing 20 sites needs a single view that surfaces exceptions: which sites have delivery variances outside normal ranges, which have active ATG alarms, which are approaching reorder levels, and which have had any anomalous level movements in the past 24 hours.
VX-Olympus’s multi-site fuel dashboard provides this view with filter capabilities: sites below 20% tank capacity (approaching reorder threshold), sites with active alarms, sites with variance flags, and historical reconciliation trend by site.
Delivery Schedule Optimization
Rather than scheduling deliveries on a calendar (every Tuesday regardless of level), VX-Olympus calculates days-to-minimum based on actual consumption rate and current level:
- Current level: 3,200 gallons
- Average daily consumption (7-day trailing average): 340 gallons/day
- Minimum operating level: 400 gallons (operational buffer)
- Calculated days to minimum: (3,200 - 400) / 340 = 8.2 days
With delivery lead time of 2 days, the optimal order date is 6.2 days from now — not the fixed calendar date that may arrive when the tank is at 60% or at 10% depending on that week’s traffic.
Regulatory Compliance Automation
Multi-site fuel operations face EPA Underground Storage Tank (UST) regulatory requirements that include monthly reconciliation documentation. Many state environmental agencies add their own reporting requirements beyond the federal minimum.
VX-Olympus automates the monthly reconciliation report:
- Total deliveries received (date, volume, tank)
- Total sales from dispenser records (imported via POS integration or manual entry)
- Net inventory change
- Variance (gallons and percentage)
- ATG alarm events during the period
- Leak detection test results
The report is generated from the continuous monitoring data — no manual data collection required at report time. For operators whose compliance documentation burden has been significant (manually pulling ATG printouts, transcribing delivery manifests, calculating reconciliation in spreadsheets), this automation is often the most immediately tangible operational value.
Implementation Approach
Deploying VX-Olympus for fuel operations across a multi-site portfolio follows a standard progression:
Phase 1 (sites 1–5): Proof of concept with high-value sites. Connect ATG systems, configure delivery tracking and reconciliation rules, establish baseline variance profiles for each site. Validate that the integration is capturing deliveries accurately and that the reconciliation calculation matches manual reconciliation.
Phase 2 (full portfolio rollout): Standardize the site configuration template from Phase 1. Roll out remaining sites with the validated configuration. The rollout is primarily connectivity work (ATG integration at each site) — the platform configuration scales from the template.
Phase 3 (optimization): Review 90-day reconciliation data across all sites. Identify outlier sites. Investigate and resolve systematic variances. Begin delivery schedule optimization using consumption-based projections rather than fixed calendars.
Conclusion
Multi-site fuel operations have always known that fuel balances rarely close perfectly. The variance has historically been treated as a cost of doing business accepted because identifying variance sources required too much effort.
Continuous IoT monitoring changes the calculus. The investigation that previously required reconstructing a month of manual ATG readings now happens automatically, in near-real-time, flagging specific events for attention while they are still current.
VX-Olympus provides the multi-site fuel monitoring layer: ATG integration, continuous level analysis, delivery reconciliation, and a centralized dashboard for portfolio-level exception management.
Talk to our team about a fuel operations monitoring deployment for your site portfolio.