PETR4 Pre-Salt FPSOs: Production Drivers Disclosed in FRE

The research question

> What production and operating-performance drivers does the FRE disclose for pre-salt FPSOs and related refining / financial context?

New FPSOs and ramp context

The filing highlights recent units and milestones as production drivers:

• Maria Quitéria
• Marechal Duque de Caxias
• Almirante Tamandaré / Búzios 7
• Sepetiba peak production

For workflow design, treat these as named assets to tag in extracted FRE text — then reconcile volumes against DFP/ITR operating notes in the same cycle.

Pre-salt production records

Disclosed operating highs and mix:

Pre-salt dominance is the structural story; FPSO names explain how capacity arrived. Models that still treat “Brazil production” as a single lever miss the FRE’s explicit mix language.

Refining utilization and capacity projects

Same-cycle refining disclosures:

• Refining utilization 93% in 2024
• RNEST Train 1 revamp: 115 → 130 kbpd
• Boaventura UPGN: 10.5 → 21 MM m³/d

Linking upstream boed to downstream utilization is standard FRE reading for integrated oil: production alone does not explain margin or product mix.

Sales mix and revenue scale

Revenue context from the same analysis pass:

• Diesel roughly 30–31% of sales revenue
• 2024 sales revenue R$490.8bn
• 1Q25 sales revenue R$123.1bn

Diesel weight is useful when reading crackspread commentary in the same filing cycle (below): product mix amplifies or dampens crack moves.

Price, cracks, and Adjusted EBITDA (2024)

Operating strength did not translate one-for-one into earnings:

• Brent average roughly −2% in 2024
• Diesel crackspread −39%
• Adjusted EBITDA excluding exclusives −11% YoY in 2024

This is why FRE production sections should be joined to price/margin language before concluding “higher boed = higher EBITDA.”

FX and financial result

Currency swings were material in the disclosed bridge:

1Q25 contrast in the same narrative: Adjusted EBITDA +8% vs 4Q24, with BRL +7.3% helping the finance result. FX direction flipped the finance line’s contribution even when operating metrics stayed strong.

Why automate this

FPSO names, boed records, utilization %, and FX bridges live in different FRE subsections. With apicvm you list and extract filings; your agent or pipeline answers production questions with citations:

export APICVM_KEY='apicvm_...'
export APICVM_URL='https://apicvm.dev'

curl -H "Authorization: Bearer $APICVM_KEY" \
  "$APICVM_URL/v1/documents?ticker=PETR4&type=FRE&year=2025"

curl -X POST -H "Authorization: Bearer $APICVM_KEY" \
  -H "Content-Type: application/json" \
  -d '{"document_id":"<DOCUMENT_ID>"}' \
  "$APICVM_URL/v1/document-text-extractions"

Example questions over extracted text:

• "Which new FPSOs are cited as production drivers?"
• "What were 2024 own and operated pre-salt production records?"
• "What refining utilization and RNEST / Boaventura capacity changes are disclosed?"
• "How did FX affect the 2024 financial result?"

Use cases

Methodology note

From Hold analysis of Petrobras public 2025 FRE operating-performance and related financial sections (analysis_results id 660). Figures can change in later filings — verify current FRE. Not investment advice.

Limitations

• Own vs operated boed are not interchangeable in models; keep the FRE labels.
• Crackspread and FX lines are cycle-specific; do not extrapolate without new filings.
• apicvm provides filings access and text extraction only — not production databases, price feeds, or investment recommendations.

Next steps

• PETR4 CAPEX business plan
• PETR4 natural gas competition
• PETR4 CVM filings
• Read the API docs