Why refinery margin is decided one unit at a time
A refinery's profitability is the cumulative result of thousands of operating decisions across interconnected units - how hard each unit is pushed, how product cuts are set, how much energy is consumed, how catalyst is managed. Each unit produces enormous volumes of process data, but the data that runs the unit is not the same as the data that manages the business. Throughput, yield and energy performance are often reported separately, days apart, and the picture of how the whole refinery is performing against plan is assembled slowly enough that the opportunity to act on it has often passed.
Good refinery operations reporting joins process, yield, energy and maintenance data into one operations view - so the bottleneck unit, the yield slipping against plan and the energy cost creeping up are all visible in time to do something about them.
The metrics that belong on a refinery operations dashboard
- Throughput - feed rate by unit, against capacity and plan
- Unit utilisation - actual against nameplate and against the constraint
- Yield by product - actual product slate versus target, with give-away
- Energy consumption - energy intensity by unit and per barrel processed
- Catalyst performance - activity and run-length against expectation
- Downtime and reliability - lost throughput by unit and cause
Monitoring energy consumption and catalyst performance
Energy is typically the largest controllable cost in a refinery after feedstock, and the most diffuse - spread across furnaces, compressors, pumps and steam systems where small inefficiencies accumulate quietly. Catalyst performance, meanwhile, sets the run-length and yield of major units and degrades on a schedule that has to be managed against turnaround timing. Reporting that trends energy intensity and catalyst activity by unit makes both of these manageable, exposing the units drifting from efficient operation before the cost compounds.
Identifying bottlenecks in processing units
A refinery, like any connected system, has a constraint - the unit that limits the throughput and margin of the whole plant. The constraint moves with feed slate, market conditions and unit availability, and managing it is the central discipline of refinery operations. A useful dashboard identifies the current constraint and shows what it is costing, so the operations team can decide whether to debottleneck it, reroute around it or accept it - rather than optimising units in isolation while the real limit sits elsewhere.
Linking operations to maintenance and downtime
Throughput and reliability are inseparable. An unplanned outage on a key unit ripples through the whole refinery, and the decision of when to take a unit down for maintenance is one of the most consequential in refinery management. Reporting that links operating data to maintenance and downtime - lost throughput by cause, reliability trends by unit, the production cost of deferred maintenance - lets the operation make these calls with the margin impact visible, and plan turnarounds around the units that are genuinely constraining performance.
Siloed unit reporting vs unified refinery operations reporting
| Aspect | Siloed unit reporting | Unified operations reporting |
|---|---|---|
| Constraint visibility | Optimised unit by unit | Plant-wide constraint made explicit |
| Yield management | Reviewed after the fact | Give-away exposed against target |
| Energy | Reported separately | Intensity trended by unit and per barrel |
| Maintenance decisions | Made without margin context | Lost-throughput cost visible |
Refinery operations reporting across processing contexts
Crude distillation
The front of the refinery, where cut points and energy use set the slate for everything downstream. Reporting that ties cut points to yield and energy is foundational.
Hydrocracking and conversion
High-pressure, high-value conversion where catalyst and energy dominate economics. Reporting that trends catalyst activity and conversion against plan protects both yield and run-length.
Reforming and treating
Units where product quality and octane economics matter most. Reporting that links severity to yield and give-away supports the daily optimisation calls.
The Power BI architecture behind refinery operations reporting
On a typical SolveBI deployment we land process-historian data, yield-accounting and oil-movement data, energy-management and CMMS maintenance data into Microsoft Fabric, then expose a single operations model through Power BI. Operations sees the throughput, constraint and yield view; energy and reliability teams see the intensity and downtime view; and management sees the margin and plan-versus-actual picture - all from one Power BI dataset, with the same numbers feeding compliance and environmental reporting.
Common mistakes in refinery operations reporting
- Unit-by-unit optimisation. Without the plant-wide constraint, local gains can leave the real limit untouched.
- Ignoring yield give-away. Looser-than-required specs leak margin every hour.
- Energy reported in isolation. Intensity per barrel by unit is where the savings actually sit.
- Maintenance without margin context. Downtime decisions need the lost-throughput cost attached.
- Slow, separate reporting. By the time the picture is assembled, the chance to act has passed.
From siloed unit reports to plant-wide margin control.
Book a free 30-minute consultation with a Microsoft-certified SolveBI consultant. We'll map your historian, yield and energy data, agree the right operations metrics, and quote a phased Power BI deployment you can budget against.
