A field engineering primer on the operating framework that anchors the 33% preventable-downtime number, the 60% vs 25% field productivity number, and the closed-loop deployments at ExxonMobil, ConocoPhillips, and Chevron. Why the term shows up in every supermajor briefing deck. Why the build is no longer multi-year. And what the four-week adoption path looks like for the operators who still run on fixed routes.
What Exception-Based Surveillance Actually Is
Exception-based surveillance (EBS) is an upstream operating framework that ranks every field action by a quantitative score derived from continuous monitoring of the data already in the historian, the SCADA, the accounting system, and the EAM. The crew does not visit wells on a calendar rotation. The crew visits wells whose signal moved.
Alvarez & Marsal published the canonical definition in 2015 in "The Advantages of Exception-Based Surveillance," a whitepaper that put a defensible number on what the operations leaders at the supermajors had been doing informally for a decade. Two numbers anchor the paper. EBS lifts a lease operator's value-added time from approximately 25% to approximately 60%. And it reduces production downtime by approximately 33% through earlier intervention on equipment degradation signals already present in the data.
Both numbers have held up across a decade of subsequent peer-reviewed deployments at the operators with the largest basin positions in North America. None of them required a sensor refresh. None of them required a multi-year data project. All of them required the operating change EBS describes: stop running the calendar route, start running the score.
This pillar page does one job. It defines what EBS is, walks the three operating levers inside the framework, sources the supermajor case studies that confirmed the math, and lays out the four-week adoption pattern that the same operating model now follows at the independent scale.
The Fixed-Route Default and Where It Leaks Value
Walk into any 500-well operation that has not adopted EBS and the operating model is recognizable in under an hour.
The pumper has a route. The route is a sequence of wells the pumper visits on a fixed cadence, typically every two to three days, in geographic order. The pumper checks each well, reads the tank, looks at the alarm panel, opens a few hatches, runs the dyno on the rod-pump well at the back of the lease, and moves to the next stop. The route does not change unless the dispatcher gets a call.
Three patterns leak value out of the fixed-route model.
Pattern one: the highest-value well in the basin does not get the most attention. A 350-bopd well drifting 8% off forecast and a 30-bopd stripper running steady get the same number of visits per week. The dollar at risk on the first well is an order of magnitude higher. The route does not see it.
Pattern two: equipment failure gets discovered hours after it happens. The signal that the ESP was going to fail was in the historian three days before the pumper opened the lease gate. Nobody scored it. The well is down 18 hours by the time the pumper finds the failed unit, and another two to four days before the workover finishes. The deferment shows up on the variance report two billing cycles later.
Pattern three: the route is busy and the dollar number does not move. The pumper drove 220 miles, opened 28 wellheads, and wrote 14 tickets. The aggregate effect on BOE produced this month is rounding noise. The day was full of work. None of it was the work that mattered.
EBS does not add work to that pumper's day. It changes the order. The route opens on the highest-score well in the basin, not the next stop on the calendar. The wells whose signal did not move overnight fall off the route. The wells whose signal did move land at the top. The crew runs more high-value visits per shift because the low-value visits are not there.
That is the whole framework, stated at the operating level. The cost of the build is the integration onto the existing data stack and the scoring layer that runs nightly. The cost of not running the framework is the value pattern above, compounded across every shift for every quarter the framework is not in place.
Get the WorkSync Field Ops Brief
Monthly read for upstream + midstream operations leaders. Case studies, benchmarks, and what's changing in the field. Unsubscribe anytime.
The Three Operating Levers Inside EBS
A&M's 2015 paper, the SPE and JPT case studies that followed, and the production-grade deployments now running on independent fleets all decompose EBS the same way. Three operating levers carry the framework. Each lever moves a different number on the operating budget.
Lever one: preventative maintenance, the 33% downtime lever. The maintenance lever inside EBS catches equipment degradation early enough to intervene at scheduled cost rather than emergency cost. The signal is already in the historian. The score ranks every artificial lift unit, separator, and surface facility by probability of intervention-worthy event in the next 7 days, multiplied by the FCF at risk on that asset. The workover crew schedules against the forecast, not against the failure. The 3 to 5x emergency-cost premium disappears. The well does not go offline. The 33% number, written up in detail in 33% of Your Production Downtime Is Preventable, is the maintenance lever specifically.
Lever two: dispatch and rerouting, the productivity lever. The dispatch lever inside EBS reorders the route every morning against the overnight score. The pumper, the field foreman, the well tester, the chemical truck, and the workover rig all see the same ranked plan in the truck cab by 6 AM. Calendar visits drop. Exception visits rise. The value-added share of the operator's day moves from 25% to 60%, with pump-by-priority scoring lifting that further as the loop closes. The full math is in 60% vs 25%: The Field Productivity Gap the Supermajors Already Closed.
Lever three: closed-loop lift and production control, the optimization lever. The third lever runs setpoint changes on artificial lift, separator pressures, and choke schedules against the live score. The supermajor case studies all carry a closed-loop layer behind the surveillance layer. ExxonMobil's 1,300-well Permian neural-network ESP deployment ran closed-loop optimization on the existing fleet and reported a 2% production uplift with no equipment changes. The score recommends the setpoint. The controller accepts or rejects. The model retrains overnight. Setpoint drift, dyno trends, and pressure-volume cycles converge to a tighter envelope week over week.
The three levers do not have to be deployed in parallel. Most independents start with lever two (the visible win on the route) and lever one (the visible win on the workover schedule), and add lever three once the data hygiene from the first two levers is durable. The four-week pilot covered in The Four-Week Pump-by-Priority Pilot wires the first two levers and lays the data foundation for the third.
The Supermajor Proof Block
The 2015 A&M numbers would be interesting on their own. The reason they are durable is the case studies that followed at the operators with the most basin position to defend.
ExxonMobil. Permian. 1,300-plus wells. 2% production uplift. No new sensors. Presented at the SPE Artificial Lift Conference in 2024, the deployment ran a neural-network closed-loop optimization on existing ESP fleet across more than 1,300 Permian wells. The 2% uplift was algorithmic, not equipment-driven. The signal was already in the historian. EBS plus closed-loop optimization, on a peer-reviewed venue, at scale.
ConocoPhillips. Norway business unit. 80% reduction in select maintenance activities. 90% in others. Published in JPT (Journal of Petroleum Technology) in 2024 as part of an IOCaaS deployment. The reductions were activity-level: scheduled maintenance tasks the model determined did not need to happen on the prior cadence because the signal said the equipment was fine. The freed crew hours went to higher-leverage work. The downstream effect on workover demand and parts inventory was reported as material.
Chevron. Kaybob Duvernay. 5% LOE reduction in year one. Also published in JPT in 2024, the Chevron case study covered a closed-loop lift control deployment. The 5% LOE reduction is significant on its own. The composition is more significant. The largest contributor was the shift from emergency to scheduled maintenance, which carried the 3 to 5x cost differential per event on the prior baseline.
Three operators. Three basins. Three independent peer-reviewed numbers. None of them was vendor marketing. None of them required a sensor refresh. All three deployed the same operating framework A&M described a decade earlier.
Why the Independents Have Not Adopted Yet
If the framework has been published since 2015 and the supermajor deployments have been peer-reviewed since 2024, why are most small-to-mid operators still running a fixed-route model?
Three friction points account for nearly all of the gap.
The data cleanup myth. The most common internal block is the IT-led data quality assessment that arrives three months into the AI conversation. The data is messy. The historian has unmapped tags. The conclusion that gets attached to the assessment, the wrong conclusion, is that the data must be cleaned before EBS can be deployed. The supermajors did not clean their data first. They scored what they had, ranked the day, and improved the data quality as a byproduct of operating on it.
The sensor refresh trap. The second block usually comes from a vendor pitch that opens on instrumentation gaps. The pitch is plausible. New sensors would help. The conclusion that gets attached, again the wrong one, is that the deployment is gated on a sensor refresh. The 33% downtime number, the 60% productivity number, and all three supermajor case studies were produced on data the operators already had. New sensors earn their place after the score is running, not before.
The dashboard-only deployment. The third block is the most expensive. An operator stands up a predictive layer that produces forecasts, the forecasts go into a dashboard, the dashboard is reviewed weekly by a planner, and the pumper runs the same fixed route on Tuesday. The forecast did not change tomorrow's work order. EBS only moves a number when the score reorders the truck cab tomorrow morning.
Strip those three friction points and the build is one week to integrate, one month to roll out, and one quarter to see the FCF number move.
What EBS Looks Like on a Tuesday Morning
The framework is not abstract. On a Tuesday morning at a 500-well operation running EBS, four things are different from the fixed-route model.
The score ranks every well by 6 AM. Every well in the fleet sits on a two-number score: probability of an intervention-worthy event in the next 7 days, and free cash flow at risk on the asset if the event happens. The score is computed overnight against the SCADA, historian, accounting, and EAM read of the prior 24 hours.
The route opens on the top of the score. The pumper opens the tablet. The first stop is the highest-scored well in the basin, not the next well on the calendar. The wells whose signal did not move overnight fall off the route. Same crew. More high-value visits per shift.
The field observation flows back the same day. When the pumper finds a backside leak on the well the score flagged, that observation flows back into the score before tomorrow's ranking runs. Voice-first capture converts the radio chatter and the handwritten ticket into structured records the model consumes. The score gets sharper every shift.
The workover is scheduled against the forecast. The workover crew is scheduled against the failure forecast, not against the failure. Parts are ordered against the forecast. The 3 to 5x emergency-cost premium disappears. The deferment never gets booked.
The supermajors made these four changes a decade ago. Independents can make them in a month.
The Four-Week Adoption Path
The shape of an EBS adoption pilot at a small-to-mid operator is straightforward. The full version is documented in The Four-Week Pump-by-Priority Pilot. The compressed version is four weeks, one signed metric, and a walk-away clause in writing.
Week 0: Sign the metric, not the software. The operations leader, the asset manager, and the controller pick one metric and write it down. For an EBS pilot, the most common Week-0 metrics are mean time from anomaly to first field response, deferred production recovered per crew shift in BOE, and emergency-versus-scheduled workover ratio on the pilot route. The threshold for "moved" is on a one-page document the controller signs. No license fee. No kill fee. The Impact Guarantee makes the pilot the experiment, not the procurement.
Week 1: Read-only integration on the stack you already own. SCADA or historian, production accounting, EAM or CMMS, GIS, and the pumper's existing field data capture. Read-only. Five to seven working days. No new sensors. No rip-and-replace.
Weeks 2-3: The score runs live, the plan hits the truck cab. The scoring loop runs nightly. Every well, every artificial lift unit, every open work order is scored on failure probability and FCF at risk. The ranked plan is published to the truck cab and the operations center by 6 AM. The superintendent adjudicates the top 20 items each morning and feeds the corrections back into the score.
Week 4: Measure, decide, sign. The chosen metric is measured against the Week-0 baseline. If it moved past the threshold, the operator signs the annual on the same Impact Guarantee terms and the pilot route expands to the full asset on a 60-to-90-day rollout. If it did not, the operator walks away with the integration documentation and the baseline data.
The deployed reference at a top-25 private producer running over 5,000 wells across the Western Anadarko, Permian, and Wyoming followed this exact shape and delivered 15%+ free cash flow uplift on the same well count, no new wells drilled, no incremental headcount.
Why This Operating Model Has Become Adoption-Urgent
EBS has been adoption-optional for a decade. It is not, anymore. Three forces are closing the window for the operators who still run on fixed routes.
The EPA OOOOb methane rule is forcing continuous monitoring across most onshore tank fleets through a 2026-2027 compliance window. Operators that put EBS in place during the OOOOb build absorb the monitoring upgrade as part of an operating change they were going to make anyway. Operators that do not, end up paying for the monitoring twice and never realize the operating leverage.
The supermajor playbook is migrating downstream into the top quartile of independents now. The 5,000-well reference deployments, the JPT case studies, and the SPE papers are in executive briefing decks. The first-mover advantage on FCF is real and measurable. The second-mover position is competing for the same crude on a higher cost structure.
The lease operator with 18 years of basin knowledge is retiring. The replacement, if hired at all, is running off a structured score by default. The operating model has to be in place before the institutional knowledge walks out.
EBS is not a vendor framework. It is the operating model the operators with the most basin position to defend have already productionized. The math is published. The case studies are peer-reviewed. The integration is one week. The pilot is four. The operating change is permanent.
The independents that adopt this year close the gap to the supermajors. The independents that wait, widen it.
Request Your Free Trial. Four-week EBS adoption pilot on the data stack you already own. One signed metric. Walk-away clause in writing.
