Note: Anonymized operational data. Selected excerpts shown; confidential details removed.
Report Snippet: Oil Refinary (Ops-only)
Report Content
Executive Summary
What we analyzed: CDU/VDU plus key interfaces (hydrotreaters, hydrogen); 250+ internal benchmarks, 130+ cross-unit interdependencies.
Headroom (based on ops data): Two-phase utilization lift on a stream-day basis: 90.9% (~74.54 kb/d) → 93.7% (~76.83 kb/d) → 96.0% (~78.72 kb/d) = +4.18 kb/d total.
- ◦Drivers: desalter contamination control, combustion/ΔT stability, APC-led crude-rate optimization, and vacuum/steam efficiency—sequenced to interdependencies (e.g., strong furnace↔column coupling). Hydrogen & downstream readiness checked to avoid shifting bottlenecks.
Indicative financial outcome:
- ◦Phase 1: ~+2.30 kb/d = $4-4.5M/yr
- ◦Phase 2: ~+1.89 kb/d = $3-3.5M/yr
- ◦Total: ~+4.18 kb/d ≈ $7-8M/yr in rate uplift only
- ◦In addition: energy savings & off-spec avoidance
Immediate ops priorities (quick-wins + control discipline):
- ◦Contamination control (desalter): deploy online salts/chlorides analyzers, set alarm bands, enforce SOP response → ≥95% hours in-spec.
- ◦Combustion & ΔT coordination: advanced combustion control; stabilize column ΔT (8–12°F → ≤3–4°F); align furnace duty to separation; manage excess air and bridgewall stability.
- ◦Cross-unit visibility: live monitors for high-impact correlations; embed alerts into shift huddles to cut daily crude CV.
Internal Best Performance Benchmarking
| Area / KPI | IBH (days) | Stability tests (illustrative) | How we use it | ||
|---|---|---|---|---|---|
| CDU/VDU throughput (kb/d) | 7–14 | Daily crude CV ≤5–7%; ≤1 unscheduled event; ΔT within spec | Sets Phase-1 rate bands and Phase-1 exit criteria | ||
| Quality (salts ptb / chlorides ppm) | 14–30 | ≥95% hours in-spec; no corrosion alarms; lab/online match | Defines analyzer alarm limits & SOP response cadence | ||
| Energy (EII / SFC / Heater efficiency) | 14–30 | Excess air in limits ≥90% hours; APC on ≥85% hours | Locks combustion/APC control bands; tracks EMS gains | ||
| Reliability (On-stream / Mech. Avail. / Unplanned) | 30–60 | MTBF ≥ P75; MTTR ≤ median; planned only | Gates Phase-2 stability & sustained high-rate days |
Principle: Targets come from Internal Best Operating Windows – plant's best stable performance under real constraints. Each lever has specific observed period and stability tests.
Primary Performance Gaps
| Process/Area | KPI | Current | Benchmark / Target | Gap | Primary Drivers |
|---|---|---|---|---|---|
| System-wide | Crude rate vs stream-day | 90.9% (~74.54 kb/d) | 93.7% (~76.83) → 96.0% (~78.72) | +2.30 → +4.18 kb/d | Heat integration; APC utilization; cross-unit coordination |
| Quality (desalter) | Salts (ptb) | 11.9 | ≤ 6.2 (≥95% hours in-spec) | +48% vs target | Lagging real-time control; inconsistent SOP/controls |
| Quality (desalter) | Chlorides (ppm) | 8.8 | ≤ 4.7 (≥95% hours in-spec) | +47% vs target | Slow alarm/response cadence; weak control discipline |
| Control stability | Daily crude CV | 15–20% | ≤5–7% (IBH 7–14d) | High variability | Manual interventions; weak furnace↔column sync |
| Energy | EII / SFC | High vs Internal Best | EII –3 to –5 points / SFC –3–5% (≈ heater efficiency +3–5 pp) | Material | Excess air; combustion tuning; waste heat; poor heat recovery |
| Reliability | On-stream / Mech. Availability / Unplanned downtime | Below vs Internal Best | On-stream +3–5 pp; Unplanned ↓30–40% | Significant | Planned/unplanned downtime; chronic rate losses |
Guardrails: Sequence execution to the data across 137 high-impact interdependencies (e.g., furnace duty → column separation (ΔT); desalter performance → downstream quality/rates; vacuum ↔ steam efficiency) so wins stack and avoid local optima. Hydrogen supply & hydrotreaters' capacity validated as rates rise.