Rules & Scoring¶
This section summarizes the 2025.1 rules and how teams are scored. Track‑specific examples and clarifications live on the Judging pages.
Prizes and recognition — updates in progress
The prize structure for the current cycle is being actively reworked and improved. In addition to the SPE/IADC paper publication opportunity, we plan to include additional forms of recognition, contingent on available budget. More updates to come.
General Rules¶
- Eligibility and team composition follow prior years; cite all external sources and code. Publishable artifacts are encouraged (see Resources → Downloads and OSDC references).
- Submissions MUST include required datasets and documentation. Use the fixed‑schema
drilling_timeseries.csv(≥1 Hz; D‑WIS names) and a static LAS 2.0 log where suggested. - Interoperability checks are REQUIRED prior to final tests. For Mode Physical (P) these are read‑only/shadow (no PLC writes). For Mode Virtual (V), verify OpenLab API connectivity.
- Safety and compliance are MANDATORY. Judges grade both the plan and implementation.
- Outcomes‑first and algorithm‑agnostic: judges score outcomes and evidence; no preference is given to specific algorithms or model families (e.g., PID vs MPC vs RL; CNN vs rules vs LLM), provided constraints and safety are met.
Mandatory Rule
- Teams MUST pass interoperability checks before Phase II testing.
- Mode P interop is read‑only/shadow; teams MUST NOT write to PLCs/actuators during checks.
- Safety compliance SHALL be demonstrated in both documentation and implementation.
Innovation Uplift (Applies in All Categories)¶
Judges may award the top of a category’s band when a documented novelty (hardware, sensing, control, HF/UX, data tooling, compression/efficiency) materially improves outcomes, safety, or clarity. This does not exceed the category maximum; it recognizes innovation anywhere, not only via bonuses. Evidence should include:
- Problem framing and design rationale
- Tests/ablations or comparative baselines showing measurable impact
- Safety compliance and operational constraints
- Reusability (artifacts, methods, or documentation)
Scoring Summary (2025.1)¶
| Category | Points |
|---|---|
| Trajectory / Test / Target Accuracy | 25 |
| Drilling Efficiency (Time to TD / ROP) | 25 |
| Well‑Control & Dysfunction Handling | 15 |
| Safety & Compliance | 10 |
| Human Factors & UI Clarity | 10 |
| Automation Quality & Innovation | 15 |
| Edge‑AI Bonus (Optional) | +10 max |
| Category | Points |
|---|---|
| Trajectory / Path Quality / Tortuosity | 20 |
| Drilling Efficiency (Time to TD / ROP) | 20 |
| Safety & Procedure Compliance (E‑Stop, Guards, Limits) | 20 |
| Data Quality (CSV Schema + LAS, Calibration, ≥1 Hz, Semantic Headers) | 10 |
| Human Factors & UI Clarity | 10 |
| Design & Automation Quality | 20 |
| Edge‑AI Bonus (Optional) | +10 max |
Mandatory Rule
Interop demo is a Pass/Fail gate before Phase II. Teams MUST pass interop to proceed.
Edge‑AI Bonus (Optional)¶
- Scored on correctness vs ground‑truth windows, grounding to CSV, latency/footprint, clarity/safety, and innovation. AI score is scaled to a maximum of +10 points.
- Default cap is 10% of total; committee may raise to ≤15%.
- Expectation: focus your effort on generating/fine-tuning data and optimizing compact multimodal models.
- Inference‑only, offline, JSON outputs (e.g., Explain‑This‑Event or Rig‑State Textualizer).
- Small‑model, edge‑deployable focus: teams may prototype with larger cloud‑hosted models during development, but final deliverables and performance are evaluated using locally run or on‑rig small models under limited inference hardware.
- Suggested caps: ≤2 vCPU or ≤20 W GPU, ≤4–8 GB RAM (non‑LLM) and ≤8 vCPU or ≤1,150 W GPU, ≤16–64 GB RAM (LLM); ≤15 min total inference runtime. Rig compute otherwise open. All AI runs must be offline during judging; no PLC/actuator writes.
- Custom task option: teams may propose a custom offline JSON task with objective ground truth and the same interface for pre‑approval (recommended deadline: 2026‑02‑01). Approved tasks use the same AI sub‑rubric and scale to +10.
See track rubrics for examples and clarifications: