#4737 Meta Prompt for Anticipating Problems During a User Specified Operation
#4737 Meta Prompt for Anticipating Problems During a User Specified Operation
#4737 Goal is Meta-prompting for:
1. User describes:
1.A. Task to be performed
1.B. Human factors
1.C. Equipment
1.D. Environment
2. LLM task: Anticipate 100 things that could go wrong
Here is a structured meta-prompt designed to force an LLM to systematically generate 100 distinct, non-redundant, and mechanistically detailed failure modes across the four input dimensions.
Meta-Prompt Template
# System: Comprehensive Failure Mode & Hazard Anticipator (CFM-HA)
## Objective
You are a principal systems safety, human factors, and reliability engineer.
Your task is to perform an exhaustive, multi-dimensional risk anticipation analysis
on an operational scenario provided by the user. You must generate exactly 100 distinct,
realistic, and consequential failure modes ("things that could go wrong").
---
## Operational Scenario Inputs
- **1.A Task to be Performed:** [Insert operational workflow, objectives, phases, and timing]
- **1.B Human Factors:** [Insert operator qualifications, fatigue levels, cognitive load, crew composition, communication channels]
- **1.C Equipment & Tools:** [Insert primary hardware, software, power sources, tooling, instrumentation, and consumables]
- **1.D Environment & Conditions:** [Insert physical surroundings, ambient variables, weather/climate, lighting, constraints, hazards]
---
## Execution Directives
1. **Exact Count:** You must output exactly 100 distinct failure modes numbered sequentially from 1 to 100.
2. **Exhaustive Decomposition Matrix:** To prevent repetition and ensure broad coverage across the entire problem space, allocate the 100 failure modes strictly across the following 10 categories (10 items per category):
- **Category 1 (Items 1–10): Task & Procedure Failures** (sequencing errors, omission, race conditions, timing mismatches, incorrect verification).
- **Category 2 (Items 11–20): Operator Cognitive & Psychomotor Failures** (tunnel vision, confirmation bias, slips, lapses, spatial disorientation, motor over-correction).
- **Category 3 (Items 21–30): Communication & Coordination Breakdowns** (misinterpreted jargon, dropped handoffs, ambiguous callouts, hierarchy gradient friction, feedback latency).
- **Category 4 (Items 31–40): Equipment Primary Functional Failures** (structural fatigue, component seizure, calibration drift, sensor bias, electrical shorts).
- **Category 5 (Items 41–50): Equipment-Human Interface (HMI) Failures** (mode confusion, parallax error, control adjacency errors, unannunciated state changes, alarm fatigue).
- **Category 6 (Items 51–60): Environmental Degradation & External Stressors** (thermal extremes, particulate contamination, moisture ingress, acoustic masking, vibration resonance).
- **Category 7 (Items 61–70): Consumables, Utilities & Auxiliary Dependencies** (power transients, contaminated fluid/fuel, depleted reserves, cable snags, tether fouling).
- **Category 8 (Items 71–80): Cross-Domain Coupling & Cascading Failures** (e.g., environmental factor induces equipment fault, which triggers human misdiagnosis).
- **Category 9 (Items 81–90): Edge Cases, Rare Transitions & Latent Conditions** (untested boundary transitions, dormant latent bugs, legacy workarounds, maintenance traps).
- **Category 10 (Items 91–100): Recovery & Secondary Response Failures** (counterproductive abort actions, incorrect manual override, collateral damage during mitigation).
3. **Format Per Item:** Each item must be formatted as:
`[Number]. **[Concise Failure Name]**: [Direct Mechanism/Trigger] -> [Operational Consequence]`
4. **Negative Constraints:**
- No generic placeholders (e.g., avoid "The user makes a mistake"; specify the exact slip/lapse).
- No duplicate failure mechanisms under different wording.
- Do not stop early or summarize mid-list.
Why this structure guarantees 100 distinct items
Partitioned Quotas: LLMs default to stopping after 15–25 items when asked open-endedly. Enforcing 10 categories with a hard ceiling of 10 items each distributes the generation evenly across orthogonal axes.
Causal Syntax (
Mechanism -> Consequence): Prevents the model from writing broad thematic statements and forces concrete mechanical descriptions.Coupling & Recovery Buckets (Categories 8 & 10): Prevents the prompt from analyzing elements in complete isolation by capturing multi-variable chain reactions and abort-handling errors.
For 1.A. thru 1.D. provide the likely data for operations of NASA Lunar Terrain Vehicles (LTVs) on the lunar surface
Here is the operational scenario data for NASA’s Artemis Lunar Terrain Vehicle (LTV) surface operations, structured across dimensions 1.A through 1.D:
1.A. Task to be Performed (Operational Concept & Workflow)
Primary Mission Modes:
Crewed Extravehicular Activity (EVA) Traverses: 2 suited crew members conducting up to 8-hour driving sorties (up to 20 km round-trip per EVA) across the Lunar South Pole.
Tele-Robotic / Autonomous Science Operations: Remote ground control from Earth/Gateway between crewed missions (spanning up to ~11 months/year).
Logistics & Pre-positioning: Transporting cargo, science tools, and volatile samples between landing sites, assets, and base camps.
Core Sub-Tasks:
Ingress/egress of suited crew, securing restraints, and stowing EVA toolkits.
In-situ geological sampling, volatile core-drilling, and cryogenic sample storage.
Sorties into and out of Permanently Shadowed Regions (PSRs) for up to 2 hours at a time.
Ascending/descending crater rims and boulder-strewn slopes up to 20°.
Payload handling and instrument deployment via a robotic manipulator arm.
Autonomous waypoint navigation and hazard avoidance (slopes, craters, boulder fields).
Battery recharge sequences via surface power stations or deployable solar arrays.
1.B. Human Factors (Crew Capabilities, Constraints & Workload)
Crew Composition: 2 astronauts in Exploration Extravehicular Mobility Unit (xEMU / Axiom xEVAS) spacesuits.
Physical & Ergonomic Constraints:
Pressurized glove rigidity reducing manual dexterity, grip strength, and fine motor tactile feedback.
Rigid spacesuit torso and helmet assembly severely restricting torso rotation, downward line-of-sight, and head azimuth/elevation.
Suited mass (~275 kg per astronaut with PLSS; 550 kg total for crew) under 1/6th gravity (0.166 g), altering center of mass and momentum dynamics.
Cognitive Load & Perception:
Visual impairment from extreme lighting: stark collimated sunlight and deep, pitch-black shadows (no atmospheric light scattering).
Loss of depth perception and contrast across wash-out terrain and shadow boundaries.
High cognitive demand managing suit consumables (O2, battery, water) while monitoring vehicle telemetry and navigational waypoints.
Communications & Latency:
UHF/Wi-Fi voice and telemetry loops between crew members, LTV, Human Landing System (HLS), and Earth (nominal 2.5–3.0 s round-trip latency to Mission Control).
Intermittent communications dropouts in deep crater interiors and PSRs requiring autonomous crew decision-making.
1.C. Equipment & Tools (Vehicle, Subsystems & Payloads)
Vehicle Mobility & Chassis:
Unpressurized, all-electric multi-wheel drive chassis with independent suspension and 4-wheel active steering.
Non-pneumatic, compliant compliant-mesh or airless tires designed for deep regolith traction.
Single-fault tolerant drive, steering, and braking architectures.
Power & Thermal Control:
High-density lithium-ion or solid-state batteries with survival heaters for lunar night/shadows (designed to survive up to 150 hours of darkness).
Articulated solar arrays and deployable radiator panels with dust-repellent coatings.
Avionics, Sensors & Perception:
Dual-redundant LiDAR, stereo visual odometry, and high-dynamic-range (HDR) cameras for terrain mapping and hazard detection.
High-power directional LED headlights and field illuminators mounted on articulated masts.
Inertial Measurement Units (IMUs), star trackers, and LunaNet / Lunar Communication Relay optical/RF transceivers.
Controls & Interfaces (HMI):
Suit-operable multi-axis hand controller / joystick designed for pressurized gloves.
High-contrast, anti-glare display panels legible in harsh direct sunlight and complete darkness.
Payloads & Manipulators:
Multi-DOF robotic manipulator arm equipped with end-effector tools, cameras, and sample containment grips.
Cryogenic sample containment boxes, rake/tongs, drive tubes, and VIPER-heritage spectrometers/neutron detectors.
250 kg payload capacity in crewed mode (up to 1,600 kg in uncrewed logistics mode).
1.D. Environment & Operating Conditions (Lunar South Pole)
Gravitational Field: 1/6 g (1.62 m/s^2), causing low ground friction, increased wheel slip, vehicle bouncing, and longer braking distances.
Atmosphere: Hard vacuum (10^-10 to 10^-12 torr); no convective cooling, requiring strict radiative/conductive thermal paths.
Thermal Extremes:
Direct sunlight: up to +120-degrees C (393 K}).
Permanently Shadowed Regions (PSRs) and lunar night: down to -180-degrees C to -246-degrees C (27 K to 93 K}).
Regolith & Geotechnical Hazards:
Ultra-fine, highly abrasive, sub-micron jagged dust particles with high electrostatic charge adhering to optical lenses, radiators, solar arrays, and mechanical seals.
Variable soil mechanics: highly compressible loose powdery dust (sinkage risk) overlaid on hard subsurface permafrost/rock.
Craters, scarps, boulder fields, and steep slope grades exceeding 20-degrees.
Radiation & Space Environment:
Galactic Cosmic Radiation (GCR), Solar Particle Events (SPE), and continuous micrometeorite bombardment without atmospheric shielding.
Illumination Geometry:
Extremely low solar elevation angles (1-degree to 3-degrees grazing angle at the South Pole), casting kilometer-long shadows and creating blinding horizontal glare when driving sunward.
For 1.A. thru 1.D. provide the likely data for operations of a extra-vehicular activity (EVA) (i.e. space walk) by astronauts on the International Space Station (ISS)
1.A. Task to be Performed (Operational Concept & Workflow)
Primary Mission Modes & Objectives:
Orbital Replacement Unit (ORU) Servicing & Upgrades: Removal and replacement (R&R) of critical external hardware (e.g., Main Bus Switching Units [MBSUs], Pump Modules [PMs], External High Definition Cameras [EHDCs]).
Power System Upgrades: Structural preparation, bracket installation, cable routing, and mechanical deployment of ISS Roll-Out Solar Arrays (iROSA).
Thermal & Fluid Loop Maintenance: Jumper reconfigurations, ammonia vent/fill operations, and leak inspections on the external Active Thermal Control System (ATCS) loops.
Science & Payload Servicing: External payload installation on the Columbus External Payload Facility (CEPA), Japanese Experiment Module – Exposed Facility (JEM-EF), or AMS-02 (Alpha Magnetic Spectrometer) repair tasks.
Core Sub-Tasks & Timelines:
Pre-EVA & Depressurization: Campout protocol or in-suit light exercise prebreathe protocol (100% O_2) to purge nitrogen, suit donning in Quest Equipment Lock, transfer to Crew Lock, and stepped airlock depressurization to vacuum.
Sortie Execution: 6.5 to 7.0 hour nominal duration outside the hatch (extendable to 8.0 hours max consumable limit).
Mobility & Translation: Translation along the Integrated Truss Structure (ITS) using handrails and the Crew and Equipment Translation Aid (CETA) cart; ingress into the Portable Foot Restraint (PFR) on the Space Station Remote Manipulator System (SSRMS / Canadarm2).
Hardware Manipulation: Fastener torquing, structural bolt release, Blind-Mate Connector mating/demating, and multi-layer insulation (MLI) blanket management.
Ingress & Repressurization: Equipment stowage, hatch closure, leak check, controlled airlock repressurization, suit doffing, and post-EVA biomedical debrief.
1.B. Human Factors (Crew Capabilities, Constraints & Workload)
Crew Composition & Roles:
2 suited EVA astronauts (Lead Spacewalker EV1 in red stripes, EV2 in pure white).
Supported by an Intravehicular Activity (IVA) crew member directing steps from the flight deck/cupola, an SSRMS robotic arm operator, and the ground EVA console (Houston MCC).
Biomechanics & Physical Demands:
Operating against a constant internal suit pressure of 4.3 psia (29.6 kPa) pure O_2, creating persistent joint resistance (especially at the elbows, shoulders, and fingers).
Severe forearm fatigue and fingernail delamination (onycholysis) from repetitive pressurized glove actuation over 7+ hours.
Zero-gravity mass handling: manipulating orbital replacement units weighing upwards of 500 to 1,000 kg where momentum and inertia must be countered purely through grip, foot restraints, and core stabilization.
Sensory, Cognitive & Visual Workload:
Visual restriction from the Extravehicular Visor Assembly (EVA) helmet, with zero torso flexion and blind spots requiring Helmet Mounted Cameras and small wrist mirrors to view chest-mounted displays.
Day/night thermal and visual transitions every 45 minutes: abrupt shifts from blinding specular solar reflections to deep orbital shadow requiring helmet-mounted incandescent/LED floodlights.
High split-attention cognitive load: tracking step-by-step cue cards, monitoring life support consumables (battery Ah, O_2 margin, CO_2 partial pressure, cooling water rate), and executing precise mechanical work.
Physiological Hazards:
Risk of Decompression Sickness (DCS / "the bends") during transition from 14.7 psia cabin atmosphere to 4.3 psia suit pressure.
Risk of water intrusion inside the helmet from Liquid Cooling and Ventilation Garment (LCVG) condensation or Sublimator loop leaks.
1.C. Equipment & Tools (Suits, Servicing Hardware & Safety Systems)
Extravehicular Mobility Unit (EMU):
Hard Upper Torso (HUT): Rigid aluminum/fiberglass shell housing shoulder/arm bearings and umbilical interfaces.
Primary Life Support System (PLSS): Closed-loop backpack containing oxygen ventilation, Water Membrane Evaporator / Sublimator, and Metal Oxide (METOX) or Lithium Hydroxide (LiOH) CO_2 scrubbers.
Display and Control Module (DCM): Chest-mounted mechanical switch and digital telemetry interface.
Liquid Cooling and Ventilation Garment (LCVG): Spandex undergarment with 84 meters of water-circulating capillary tubing and air ventilation ducts for core thermoregulation.
Emergency Rescue Hardware:
Simplified Aid for EVA Rescue (SAFER): Latch-on cold-gas (N_2) thruster backpack providing self-rescue capability (3.05 m/s total delta-V) if an astronaut becomes untethered from the station.
Tethers & Restraints:
85-foot Safety Tethers with dynamic take-up reels.
Dual-hook retractable Waist Tethers, adjustable local work tethers, and Portable Foot Restraints (PFR / APFR).
EVA Specialized Power & Hand Tools:
Pistol Grip Tool (PGT): Microprocessor-controlled, battery-powered cordless drill/driver with calibrated torque (0.5 to 38 N cdot m), angle, and turn limits.
Handheld Mechanics Tools: Trash bags, soft/hard equipment caddies, ratchets, wire cutters, torque multipliers, vice grips, and scraper tools.
Fluid Handling: Fluid quick disconnect (QD) tools, ammonia vent jumpers, and Toxic Vapor Purge bags.
1.D. Environment & Operating Conditions (Low Earth Orbit / ISS Microgravity)
Orbital Mechanics & Dynamics:
Microgravity (sim 10^-6 g environment) at an orbital altitude of sim 400 to 420 km, moving at an orbital velocity of ~ 7.66 km/s (27,600 km/h).
90-to-92-minute orbital period (45 minutes direct solar irradiation alternating with 45 minutes of complete Earth-eclipse night).
Atmospheric State:
Ultra-high vacuum (10^-7 to 10^-9 torr), characterized by Atomic Oxygen (AO) erosion that oxidizes external coatings and polymers over long durations.
Thermal Extremes:
External structural surface temperatures ranging from -150-degrees C (-238-degrees F) during orbital night to +120-degrees C (+248-degrees F) in direct sunlight.
Space Debris & Radiation Environment:
Micrometeoroid and Orbital Debris (MMOD) hypervelocity impact threats (velocities up to 10 to 15 km/s) mitigated by the Thermal Micrometeoroid Garment (TMG) multi-layer Ortho-Fabric exterior.
Direct ionizing radiation, Galactic Cosmic Rays (GCR), and passage through the South Atlantic Anomaly (SAA) elevating trapped proton exposure.
Contamination Hazards:
Toxic anhydrous ammonia (NH_3) venting or residual crystals from external cooling loops adhering to suits, requiring solar bake-out protocols before airlock re-entry.
Plume impingement from visiting vehicle thrusters or ISS attitude control thrusters.
Provide a generalized Python application using Google Gemini API to provide a system prompt followed by a user prompt that includes the user-specified information (i.e. task, human, equipment, environment). Write the system instructions, user-specified information and response to a markdown file to be saved in the specified directory.
A modular Python script implementing the workflow using the official google-genai SDK is provided below. It structures the system instructions, injects the four-part operational scenario into the user prompt, sends the request to the Gemini model, and writes a complete Markdown artifact to a specified target directory.
Prerequisites
Install the official Google GenAI SDK:
pip install google-genai
Make sure your API key is exported in your environment:
export GEMINI_API_KEY="your-api-key-here"
Python Application (hazard_anticipator.py)
import os
import sys
from datetime import datetime
from pathlib import Path
from typing import Optional
from google import genai
from google.genai import types
# ==============================================================================
# 1. SYSTEM INSTRUCTION (META-PROMPT)
# ==============================================================================
DEFAULT_SYSTEM_INSTRUCTION = """\
# Role & Purpose
You are an expert Systems Safety, Human Factors, and Reliability Engineer.
Your objective is to conduct an exhaustive, rigorous risk anticipation analysis on the provided operational scenario.
# Task Directives
1. Generate exactly 100 distinct, realistic, and consequential failure modes / things that could go wrong.
2. Structure the 100 items evenly across the following 10 categories (10 items per category):
- Category 1 (Items 1-10): Task & Procedure Failures (sequencing, omissions, timing errors)
- Category 2 (Items 11-20): Operator Cognitive & Psychomotor Failures (slips, lapses, biases, spatial disorientation)
- Category 3 (Items 21-30): Communication & Coordination Breakdowns (misinterpreted callouts, handoff drops, telemetry latency)
- Category 4 (Items 31-40): Equipment Primary Functional Failures (mechanical seizure, sensor bias, electrical shorts)
- Category 5 (Items 41-50): Equipment-Human Interface (HMI) Failures (mode confusion, unannunciated state changes, alarm fatigue)
- Category 6 (Items 51-60): Environmental Degradation & External Stressors (thermal extremes, particulate/dust, vacuum/pressure)
- Category 7 (Items 61-70): Consumables, Utilities & Auxiliary Dependencies (power transients, depleted margins, fluid line fouling)
- Category 8 (Items 71-80): Cross-Domain Coupling & Cascading Failures (environmental trigger -> equipment fault -> human misdiagnosis)
- Category 9 (Items 81-90): Edge Cases, Boundary Transitions & Latent Conditions (dormant bugs, unverified edge states)
- Category 10 (Items 91-100): Recovery & Secondary Response Failures (counterproductive aborts, improper manual overrides)
3. Format each item strictly as:
`[Number]. **[Concise Failure Title]**: [Mechanism / Causal Trigger] -> [Operational Consequence]`
4. Avoid generic placeholders or duplicates. Maintain technical precision throughout.
"""
# ==============================================================================
# 2. RUNNER & ARTIFACT GENERATOR
# ==============================================================================
class ScenarioAnticipatorApp:
def __init__(
self,
api_key: Optional[str] = None,
model_name: str = "gemini-3.5-flash-lite",
):
"""Initializes the Google GenAI client and configuration."""
#self.api_key = api_key or os.environ.get("GEMINI_API_KEY")
#if not self.api_key:
# raise ValueError(
# "Gemini API key not found. Set the GEMINI_API_KEY environment variable "
# "or pass it explicitly."
# )
self.client = genai.Client(api_key="....................")
self.model_name = model_name
def format_user_prompt(
self,
task: str,
human_factors: str,
equipment: str,
environment: str,
scenario_title: str = "Operational Scenario",
) -> str:
"""Formats the 4 user-specified dimensions into a structured prompt."""
return f"""\
# Operational Scenario: {scenario_title}
### 1.A. Task to be Performed
{task.strip()}
### 1.B. Human Factors
{human_factors.strip()}
### 1.C. Equipment & Tools
{equipment.strip()}
### 1.D. Environment & Operating Conditions
{environment.strip()}
---
Please perform the exhaustive 100-point failure mode anticipation analysis following the system instructions.
"""
def generate_analysis(
self,
user_prompt: str,
system_instruction: str = DEFAULT_SYSTEM_INSTRUCTION,
temperature: float = 0.4,
) -> str:
"""Invokes the Gemini model with system instructions and user prompt."""
config = types.GenerateContentConfig(
system_instruction=system_instruction,
temperature=temperature,
)
response = self.client.models.generate_content(
model=self.model_name,
contents=user_prompt,
config=config,
)
return response.text
def save_artifact(
self,
output_dir: str | Path,
file_name: str,
scenario_title: str,
system_instruction: str,
user_prompt: str,
response_text: str,
) -> Path:
"""Writes the inputs, system instructions, and LLM output to a markdown file."""
target_path = Path(output_dir)
target_path.mkdir(parents=True, exist_ok=True)
if not file_name.endswith(".md"):
file_name += ".md"
full_file_path = target_path / file_name
timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
markdown_content = f"""# Systems Hazard & Failure Analysis Report
**Scenario:** {scenario_title}
**Generated On:** {timestamp}
**Model:** `{self.model_name}`
---
## 1. System Prompt (Engineering Directives)
```markdown
{system_instruction.strip()}
2. Input Operational Context (User-Specified)
{user_prompt.strip()}
3. Analysis Response (100 Anticipated Failure Modes)
{response_text.strip()}
"""
with open(full_file_path, "w", encoding="utf-8") as f:
f.write(markdown_content)
return full_file_path
==============================================================================
3. EXAMPLE EXECUTION PIPELINE
==============================================================================
if name == "main":
# Define destination directory
OUTPUT_DIRECTORY = "./risk_analysis_reports"
REPORT_FILENAME = "LTV_SouthPole_Hazard_Report.md"
# Define the 4 input sections (Example: NASA LTV Lunar Surface Sortie)
task_description = """
Conduct an 8-hour, 2-person crewed sortie traversing up to 15 km from the lander
at the Lunar South Pole. Mission includes descending a 15-degree crater rim,
ingress into a Permanently Shadowed Region (PSR) for 90 minutes, drilling a 1-meter
cryogenic core sample, sealing the volatile canister, and returning to base.
"""
human_factors = """
2 crew members in pressurized xEVAS spacesuits (mass ~275 kg each with PLSS).
Reduced tactile sensitivity through pressurized gloves, limited helmet field-of-view,
operating at hour 5 of an EVA with accumulating forearm fatigue. High cognitive load
monitoring suit consumables alongside vehicle telemetry and navigation path.
"""
equipment_data = """
4-wheel independent electric drive Lunar Terrain Vehicle (LTV) with non-pneumatic
compliant-mesh tires, mast-mounted high-dynamic-range stereo cameras and LiDAR,
articulated LED headlights, mechanical rotary drill payload, and cryogenic sample box.
Suit-operable joystick hand controller.
"""
environment_conditions = """
Lunar South Pole: 1/6th gravity (1.62 m/s²), hard vacuum (1e-11 torr), ambient surface
temperatures dropping to 40 K (-233 °C) inside the PSR. Sub-micron jagged, electrostatically
charged abrasive regolith. Grazing solar angles (1-3 degrees) creating long stark shadows
and blinding horizontal glare outside the crater.
"""
# Initialize and execute
app = ScenarioAnticipatorApp(model_name="gemini-3.5-flash-lite")
print("[*] Formatting operational scenario inputs...")
scenario_title = "Lunar Terrain Vehicle (LTV) PSR Core Sampling Sortie"
user_prompt = app.format_user_prompt(
task=task_description,
human_factors=human_factors,
equipment=equipment_data,
environment=environment_conditions,
scenario_title=scenario_title,
)
print(f"[*] Querying {app.model_name} with system instructions and user inputs...")
llm_response = app.generate_analysis(
user_prompt=user_prompt,
system_instruction=DEFAULT_SYSTEM_INSTRUCTION,
)
print("[*] Saving structured report to disk...")
saved_file = app.save_artifact(
output_dir=OUTPUT_DIRECTORY,
file_name=REPORT_FILENAME,
scenario_title=scenario_title,
system_instruction=DEFAULT_SYSTEM_INSTRUCTION,
user_prompt=user_prompt,
response_text=llm_response,
)
print(f"[+] Successfully generated and saved markdown report to:\n {saved_file.resolve()}")
Add saving a JSON file with the 100 entries so that it can be post-processed
import json
import os
from datetime import datetime
from pathlib import Path
from typing import List, Optional
from pydantic import BaseModel, Field
from google import genai
from google.genai import types
# ==============================================================================
# 1. PYDANTIC SCHEMAS FOR STRUCTURED EXTRACTION & POST-PROCESSING
# ==============================================================================
class FailureModeItem(BaseModel):
item_number: int = Field(description="Sequential ID from 1 to 100.")
category_id: int = Field(description="Category number from 1 to 10.")
category_name: str = Field(description="Name of the assigned risk category.")
failure_name: str = Field(description="Short, concise technical failure title.")
trigger_mechanism: str = Field(description="Direct physical, cognitive, or software causal trigger.")
operational_consequence: str = Field(description="Resulting downstream impact on safety or mission success.")
class HazardAnalysisResult(BaseModel):
scenario_title: str = Field(description="Title of the operational scenario evaluated.")
total_items: int = Field(description="Total count of failure modes, exactly 100.")
failure_modes: List[FailureModeItem] = Field(description="List of exactly 100 failure mode objects.")
# ==============================================================================
# 2. SYSTEM INSTRUCTION (ENGINEERING META-PROMPT)
# ==============================================================================
DEFAULT_SYSTEM_INSTRUCTION = """\
# Role & Purpose
You are an expert Systems Safety, Human Factors, and Reliability Engineer.
Your objective is to conduct an exhaustive, rigorous risk anticipation analysis on the provided operational scenario.
# Task Directives
1. Generate exactly 100 distinct, realistic, and consequential failure modes across the provided scenario.
2. Evenly allocate the 100 failure modes across these 10 distinct categories (10 items per category):
- Category 1 (Items 1-10): Task & Procedure Failures (sequencing, omissions, timing errors)
- Category 2 (Items 11-20): Operator Cognitive & Psychomotor Failures (slips, lapses, biases, spatial disorientation)
- Category 3 (Items 21-30): Communication & Coordination Breakdowns (misinterpreted callouts, handoff drops, telemetry latency)
- Category 4 (Items 31-40): Equipment Primary Functional Failures (mechanical seizure, sensor bias, electrical shorts)
- Category 5 (Items 41-50): Equipment-Human Interface (HMI) Failures (mode confusion, unannunciated state changes, alarm fatigue)
- Category 6 (Items 51-60): Environmental Degradation & External Stressors (thermal extremes, particulate/dust, vacuum/pressure)
- Category 7 (Items 61-70): Consumables, Utilities & Auxiliary Dependencies (power transients, depleted margins, fluid line fouling)
- Category 8 (Items 71-80): Cross-Domain Coupling & Cascading Failures (environmental trigger -> equipment fault -> human misdiagnosis)
- Category 9 (Items 81-90): Edge Cases, Boundary Transitions & Latent Conditions (dormant bugs, unverified edge states)
- Category 10 (Items 91-100): Recovery & Secondary Response Failures (counterproductive aborts, improper manual overrides)
3. Avoid generic placeholders or duplicate mechanisms. Maintain high technical fidelity.
"""
# ==============================================================================
# 3. CORE APPLICATION RUNNER
# ==============================================================================
class ScenarioAnticipatorApp:
def __init__(
self,
api_key: Optional[str] = None,
model_name: str = "gemini-3.5-flash-lite",
):
"""Initializes the Google GenAI client."""
self.api_key = api_key or os.environ.get("GEMINI_API_KEY")
if not self.api_key:
raise ValueError(
"Gemini API key not found. Set the GEMINI_API_KEY environment variable "
"or pass it explicitly."
)
self.client = genai.Client(api_key=".......................")
self.model_name = model_name
def format_user_prompt(
self,
task: str,
human_factors: str,
equipment: str,
environment: str,
scenario_title: str = "Operational Scenario",
) -> str:
"""Formats the 4 user-specified dimensions into a structured prompt."""
return f"""\
# Operational Scenario: {scenario_title}
### 1.A. Task to be Performed
{task.strip()}
### 1.B. Human Factors
{human_factors.strip()}
### 1.C. Equipment & Tools
{equipment.strip()}
### 1.D. Environment & Operating Conditions
{environment.strip()}
---
Execute the 100-point failure mode anticipation analysis according to your system directives.
"""
def generate_analysis(
self,
user_prompt: str,
system_instruction: str = DEFAULT_SYSTEM_INSTRUCTION,
temperature: float = 0.4,
) -> HazardAnalysisResult:
"""Invokes the Gemini model with structured schema constraints."""
config = types.GenerateContentConfig(
system_instruction=system_instruction,
temperature=temperature,
response_mime_type="application/json",
response_schema=HazardAnalysisResult,
)
response = self.client.models.generate_content(
model=self.model_name,
contents=user_prompt,
config=config,
)
# Parse output directly into Pydantic model
return HazardAnalysisResult.model_validate_json(response.text)
def save_artifacts(
self,
output_dir: str | Path,
base_filename: str,
system_instruction: str,
user_prompt: str,
analysis_data: HazardAnalysisResult,
) -> tuple[Path, Path]:
"""Saves both the full Markdown report and the raw JSON dataset."""
target_path = Path(output_dir)
target_path.mkdir(parents=True, exist_ok=True)
# Strip any extension if provided
clean_name = Path(base_filename).stem
md_file_path = target_path / f"{clean_name}.md"
json_file_path = target_path / f"{clean_name}.json"
timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
# 1. Save JSON Dataset (with metadata for downstream pipelines)
json_payload = {
"metadata": {
"scenario_title": analysis_data.scenario_title,
"model": self.model_name,
"generated_at": timestamp,
"total_items": len(analysis_data.failure_modes),
},
"failure_modes": [item.model_dump() for item in analysis_data.failure_modes],
}
with open(json_file_path, "w", encoding="utf-8") as f:
json.dump(json_payload, f, indent=2)
# 2. Build and Save Markdown Report
markdown_lines = [
f"# Systems Hazard & Failure Mode Analysis Report",
f"**Scenario:** {analysis_data.scenario_title} ",
f"**Generated On:** {timestamp} ",
f"**Model:** `{self.model_name}` ",
f"**Total Anticipated Items:** {len(analysis_data.failure_modes)} ",
"\n---\n",
"## 1. System Prompt (Engineering Directives)",
"```markdown",
system_instruction.strip(),
"```",
"\n---\n",
"## 2. Input Operational Context",
user_prompt.strip(),
"\n---\n",
"## 3. Anticipated Failure Modes (100 Items)",
]
current_category = None
for item in analysis_data.failure_modes:
if item.category_name != current_category:
current_category = item.category_name
markdown_lines.append(f"\n### Category {item.category_id}: {current_category}")
entry = f"{item.item_number}. **{item.failure_name}**: {item.trigger_mechanism} -> {item.operational_consequence}"
markdown_lines.append(entry)
with open(md_file_path, "w", encoding="utf-8") as f:
f.write("\n".join(markdown_lines))
return md_file_path, json_file_path
# ==============================================================================
# 4. EXECUTION PIPELINE
# ==============================================================================
if __name__ == "__main__":
OUTPUT_DIRECTORY = "./risk_analysis_output"
FILE_BASE_NAME = "LTV_SouthPole_PSR_Analysis"
# Operational Context Definition
task_description = """
Conduct an 8-hour, 2-person crewed sortie traversing up to 15 km from the lander
at the Lunar South Pole. Mission includes descending a 15-degree crater rim,
ingress into a Permanently Shadowed Region (PSR) for 90 minutes, drilling a 1-meter
cryogenic core sample, sealing the volatile canister, and returning to base.
"""
human_factors = """
2 crew members in pressurized xEVAS spacesuits (mass ~275 kg each with PLSS).
Reduced tactile sensitivity through pressurized gloves, limited helmet field-of-view,
operating at hour 5 of an EVA with accumulating forearm fatigue. High cognitive load
monitoring suit consumables alongside vehicle telemetry and navigation path.
"""
equipment_data = """
4-wheel independent electric drive Lunar Terrain Vehicle (LTV) with non-pneumatic
compliant-mesh tires, mast-mounted high-dynamic-range stereo cameras and LiDAR,
articulated LED headlights, mechanical rotary drill payload, and cryogenic sample box.
Suit-operable joystick hand controller.
"""
environment_conditions = """
Lunar South Pole: 1/6th gravity (1.62 m/s²), hard vacuum (1e-11 torr), ambient surface
temperatures dropping to 40 K (-233 °C) inside the PSR. Sub-micron jagged, electrostatically
charged abrasive regolith. Grazing solar angles (1-3 degrees) creating long stark shadows
and blinding horizontal glare outside the crater.
"""
# Run workflow
app = ScenarioAnticipatorApp(model_name="gemini-3.5-flash-lite")
scenario_title = "Lunar Terrain Vehicle (LTV) PSR Core Sampling Sortie"
print("[*] Formatting operational scenario inputs...")
user_prompt = app.format_user_prompt(
task=task_description,
human_factors=human_factors,
equipment=equipment_data,
environment=environment_conditions,
scenario_title=scenario_title,
)
print(f"[*] Querying {app.model_name} with Pydantic schema enforcement...")
result_data = app.generate_analysis(
user_prompt=user_prompt,
system_instruction=DEFAULT_SYSTEM_INSTRUCTION,
)
print(f"[*] Successfully parsed {len(result_data.failure_modes)} items.")
print("[*] Writing Markdown report and JSON dataset to disk...")
md_path, json_path = app.save_artifacts(
output_dir=OUTPUT_DIRECTORY,
base_filename=FILE_BASE_NAME,
system_instruction=DEFAULT_SYSTEM_INSTRUCTION,
user_prompt=user_prompt,
analysis_data=result_data,
)
print(f"[+] Markdown Report: {md_path.resolve()}")
print(f"[+] Post-Processable JSON: {json_path.resolve()}")
Output JSON Format Sample
The generated .json file conforms to the following schema:
{
"metadata": {
"scenario_title": "Lunar Terrain Vehicle (LTV) PSR Core Sampling Sortie",
"model": "gemini-2.5-flash",
"generated_at": "2026-08-30 14:02:26",
"total_items": 100
},
"failure_modes": [
{
"item_number": 1,
"category_id": 1,
"category_name": "Task & Procedure Failures",
"failure_name": "Canister Seal Before Thermal Equalization",
"trigger_mechanism": "Crew seals the volatile canister immediately upon extraction before verifying pressure relief venting.",
"operational_consequence": "Trapped volatile sublimation overpressurizes the core tube, causing catastrophic seal failure during transport."
},
{
"item_number": 51,
"category_id": 6,
"category_name": "Environmental Degradation & External Stressors",
"failure_name": "Regolith Dust Coating on Radiator Fins",
"trigger_mechanism": "Electrostatic dust kicked up during crater descent adheres to radiator surfaces inside the shadow.",
"operational_consequence": "Radiator thermal emissivity is degraded by 60%, resulting in battery overheating upon egress into direct sunlight."
}
]
}From <https://gemini.google.com/app/78f34bb935bde72f> Google Gemini (3.7 Flash)
From <Anaconda 3, Jupyter Notebook: "What_Can_Go_Wrong.ipynb" run> using Example Pipeline
# Systems Hazard & Failure Analysis Report
**Scenario:** Lunar Terrain Vehicle (LTV) PSR Core Sampling Sortie
**Generated On:** 2026-08-30 13:09:06
**Model:** `gemini-3.5-flash-lite`
---
## 1. System Prompt (Engineering Directives)
```markdown
# Role & Purpose
You are an expert Systems Safety, Human Factors, and Reliability Engineer.
Your objective is to conduct an exhaustive, rigorous risk anticipation analysis on the provided operational scenario.
# Task Directives
1. Generate exactly 100 distinct, realistic, and consequential failure modes / things that could go wrong.
2. Structure the 100 items evenly across the following 10 categories (10 items per category):
- Category 1 (Items 1-10): Task & Procedure Failures (sequencing, omissions, timing errors)
- Category 2 (Items 11-20): Operator Cognitive & Psychomotor Failures (slips, lapses, biases, spatial disorientation)
- Category 3 (Items 21-30): Communication & Coordination Breakdowns (misinterpreted callouts, handoff drops, telemetry latency)
- Category 4 (Items 31-40): Equipment Primary Functional Failures (mechanical seizure, sensor bias, electrical shorts)
- Category 5 (Items 41-50): Equipment-Human Interface (HMI) Failures (mode confusion, unannunciated state changes, alarm fatigue)
- Category 6 (Items 51-60): Environmental Degradation & External Stressors (thermal extremes, particulate/dust, vacuum/pressure)
- Category 7 (Items 61-70): Consumables, Utilities & Auxiliary Dependencies (power transients, depleted margins, fluid line fouling)
- Category 8 (Items 71-80): Cross-Domain Coupling & Cascading Failures (environmental trigger -> equipment fault -> human misdiagnosis)
- Category 9 (Items 81-90): Edge Cases, Boundary Transitions & Latent Conditions (dormant bugs, unverified edge states)
- Category 10 (Items 91-100): Recovery & Secondary Response Failures (counterproductive aborts, improper manual overrides)
3. Format each item strictly as:
`[Number]. **[Concise Failure Title]**: [Mechanism / Causal Trigger] -> [Operational Consequence]`
4. Avoid generic placeholders or duplicates. Maintain technical precision throughout.
2. Input Operational Context (User-Specified)
# Operational Scenario: Lunar Terrain Vehicle (LTV) PSR Core Sampling Sortie
### 1.A. Task to be Performed
Conduct an 8-hour, 2-person crewed sortie traversing up to 15 km from the lander
at the Lunar South Pole. Mission includes descending a 15-degree crater rim,
ingress into a Permanently Shadowed Region (PSR) for 90 minutes, drilling a 1-meter
cryogenic core sample, sealing the volatile canister, and returning to base.
### 1.B. Human Factors
2 crew members in pressurized xEVAS spacesuits (mass ~275 kg each with PLSS).
Reduced tactile sensitivity through pressurized gloves, limited helmet field-of-view,
operating at hour 5 of an EVA with accumulating forearm fatigue. High cognitive load
monitoring suit consumables alongside vehicle telemetry and navigation path.
### 1.C. Equipment & Tools
4-wheel independent electric drive Lunar Terrain Vehicle (LTV) with non-pneumatic
compliant-mesh tires, mast-mounted high-dynamic-range stereo cameras and LiDAR,
articulated LED headlights, mechanical rotary drill payload, and cryogenic sample box.
Suit-operable joystick hand controller.
### 1.D. Environment & Operating Conditions
Lunar South Pole: 1/6th gravity (1.62 m/s²), hard vacuum (1e-11 torr), ambient surface
temperatures dropping to 40 K (-233 °C) inside the PSR. Sub-micron jagged, electrostatically
charged abrasive regolith. Grazing solar angles (1-3 degrees) creating long stark shadows
and blinding horizontal glare outside the crater.
---
Please perform the exhaustive 100-point failure mode anticipation analysis following the system instructions.
3. Analysis Response (100 Anticipated Failure Modes)
### Category 1: Task & Procedure Failures (1-10)
1. **[Premature Crater Descent Initiation]**: Crew initiates 15-degree crater rim descent before verifying LTV auxiliary braking telemetry -> Uncontrolled vehicle slide into PSR with potential rollover.
2. **[Drill Sequence Timing Omission]**: Operator bypasses thermal stabilization dwell step for the cryogenic drill payload -> Bit seizes in volatile-rich permafrost due to thermal shock.
3. **[Canister Sealing Premature Actuation]**: Crew commands cryogenic sample box closure before core sample is fully seated -> Sample contamination and loss of volatile seal integrity.
4. **[Consumable Budgeting Sequencing Error]**: Crew completes navigation waypoint checklist out of order, delaying PSR exit -> PLSS oxygen reserves approach contingency warning threshold before egress.
5. **[Sample Labeling Omission]**: Operator fails to log core sample barcode into mission database upon retrieval -> Chain of custody lost for astrobiological volatiles.
6. **[Mast Calibration Omission]**: Crew skips LiDAR/stereo camera calibration routine prior to crater descent -> Navigation system builds degraded 3D point cloud, misjudging boulder heights.
7. **[Thermal Protection Sequencing Delay]**: LTV insulation blankets deployed too late during PSR ingress -> Battery pack temperatures drop below minimum operational threshold.
8. **[Over-Torque Procedure Bypass]**: Operator ignores drill resistance warning and manually overrides torque limit -> Shear pin failure on auger shaft inside cryogenic core.
9. **[Comm-Check Omission Post-Ingress]**: Crew fails to execute baseline signal strength check upon entering PSR shadow -> Delayed discovery of dead zone communication drop.
10. **[Stowage Latch Verification Failure]**: Crew neglects secondary mechanical pin on cryogenic sample box latch -> Severe regolith vibration unlatches box during transit back to lander.
---
### Category 2: Operator Cognitive & Psychomotor Failures (11-20)
11. **[Spatial Disorientation in Glare]**: Operator experiences flash blinding from low-angle solar glare when exiting PSR -> Incorrect steering input causes LTV wheel to wedge against boulder.
12. **[Glove Tactile Slip]**: Pressurized xEVAS glove stiffness causes crew member to slip finger off joystick safety deadman switch -> LTV abruptly halts, inducing whiplash within suit.
13. **[Confirmation Bias on Telemetry]**: Operator fixates on nominal battery voltage while ignoring rising motor controller temperature warnings -> Thermal runaway forces emergency inverter shutdown.
14. **[Loss of Situational Awareness (Tunnel Vision)]**: Crew becomes hyper-focused on core drilling depth gauge, failing to monitor suit cooling water temperature -> Sub-freezing sublimator performance goes unnoticed until crew experiences mild heat stress.
15. **[Depth Perception Error in Shadows]**: Harsh, pitch-black crater shadows cause crew to misjudge regolith slope transition -> LTV drops into unseen 0.5-meter thermal contraction crack.
16. **[Fatigue-Induced Memory Lapse]**: At hour 7 of EVA, crew member forgets secondary communications frequency switchover protocol -> Loss of direct lander comms channel for 15 minutes.
17. **[Vibration-Induced Psychomotor Tremor]**: Extended operation of high-frequency drill induces hand-arm vibration syndrome symptoms in operator -> Inability to fine-tune joystick steering inputs.
18. **[Parallax Error in Stereo Display]**: Operator misinterprets helmet HUD perspective during close-quarters parking near sample site -> LTV bumper impacts lander descent stage support leg.
19. **[Mode Habituation Transfer Error]**: Crew member instinctively applies terrestrial driving reflexes (slamming brakes on skid) during lunar descent -> Loss of wheel traction control on loose regolith.
20. **[Attention Allocation Collapse]**: Crew attempts simultaneous navigation, voice reporting, and drill parameter adjustment, exceeding cognitive bandwidth -> Misses critical low-pressure suit alarm.
---
### Category 3: Communication & Coordination Breakdowns (21-30)
21. **[Relay Latency Misinterpretation]**: 2.5-second round-trip relay delay causes crew and backroom Capcom to talk over each other during hazard avoidance maneuver -> Conflicting commands executed.
22. **[Handoff Drop off Crater Rim]**: Direct line-of-sight voice comms drop out as LTV drops behind crater rim before relay satellite acquisition -> Crew and support team experience total comms blackout.
23. **[Ambiguous Callout Nomenclature]**: Crew member calls "stop" without specifying whether to halt vehicle or drill rotation -> Co-worker stops drilling while vehicle continues creeping forward.
24. **[Telemetry Data Stream Desynchronization]**: Ground control views stale LTV position data due to buffer backlog -> Capcom directs crew toward hazard already cleared by vehicle movement.
25. **[Non-Verbal Signal Misinterpretation]**: Helmet-mounted camera angle misalignment prevents co-worker from reading hand gestures in low light -> Crew member misinterprets warning wave as confirmation to proceed.
26. **[Audio Channel Cross-Talk]**: Suit intercom picks up open mic noise from secondary telemetry loop, drowning out warning tones -> Crew misses primary oxygen pressure drop alert.
27. **[Incomplete Status Handoff]**: Exiting crew member fails to verbally confirm sample box seal verification to partner -> Partner assumes box is secure and starts high-speed transit.
28. **[Jargon Misunderstanding in High Stress]**: Stressed operator uses non-standard terminology for drill motor stall -> Capcom misinterprets status, advising incorrect recovery procedure.
29. **[Autonomous Relay Point Drop]**: LTV automated relay node fails to deploy at crater rim rim-edge, severing data link between deep PSR and surface base -> Complete telemetry blackout inside crater.
30. **[Phonetic Confusion in Voice Command]**: Automated voice-assist system misinterprets "lights high" as "drive blind," disabling mast illumination inside PSR -> Total darkness envelops crew.
---
### Category 4: Equipment Primary Functional Failures (31-40)
31. **[Compliant-Mesh Tire Tread Fracture]**: Sharp, jagged basaltic rock shears Nitinol mesh strand on front-left LTV wheel -> Loss of localized tire compliance and structural integrity.
32. **[Drive Motor Harmonic Seizure]**: Extreme 40 K cold soak inside PSR causes differential thermal contraction in harmonic drive gearbox -> Front-right wheel locks up completely.
33. **[LiDAR Sensor Bias Drift]**: Severe thermal gradient across LiDAR optical window causes ranging calibration drift -> System reports open space while vehicle approaches steep drop-off.
34. **[LED Headlight Ballast Failure]**: Articulated LED array driver circuit fails due to internal solder joint micro-fracture from vibration -> Sudden loss of illumination inside PSR.
35. **[Rotary Drill Feed Mechanism Jam]**: Fine, electrostatically charged regolith infiltrates lead screw threads of drill feed mechanism -> Drill carriage seizes mid-stroke at 0.5-meter depth.
36. **[Cryogenic Seal Outgassing]**: Elastomer seal on sample box embrittles at 40 K and loses sealing preload -> Lunar volatiles outgas into vacuum before return transit.
37. **[Inverter Power Stage Blowout]**: MOSFET failure in wheel controller power inverter due to high-torque transient during crater climb -> Loss of drive power on rear axle.
38. **[Mast Pan-Tilt Actuator Failure]**: Internal gear strip in stereo camera mast gimbal prevents horizon tracking -> Camera points fixed at regolith floor during transit.
39. **[Battery Thermal Blanket Short]**: Internal heating element short-circuits against carbon fiber structural frame -> Localized battery pack overheating and safety breaker trip.
40. **[Joystick Potentiometer Degradation]**: Regolith dust migration past joystick boot causes resistive track contamination -> Erratic steering inputs and uncommanded vehicle drift.
---
### Category 5: Equipment-Human Interface (HMI) Failures (41-50)
41. **[Mode Confusion on Drive Controller]**: Operator inadvertently switches LTV from Crab Steering mode to Ackerman mode during tight turn -> Vehicle trajectory diverges unexpectedly toward crater wall.
42. **[Unannunciated Autonomy State Change]**: LTV switches from manual assist to autonomous hazard avoidance without clear auditory cue -> Crew fights joystick inputs, destabilizing vehicle.
43. **[Alarm Fatigue Desensitization]**: Continuous low-level caution tones for minor battery thermal deltas cause crew to ignore high-priority drill motor overheat alarm -> Permanent motor burnout.
44. **[HUD High-Glare Washout]**: Horizontal solar rays wash out helmet visor HUD telemetry display -> Crew cannot read suit oxygen pressure or vehicle speed without shading visor manually.
45. **[Haptic Feedback Failure]**: Force-feedback actuator in suit-operable joystick fails, leaving control completely limp -> Operator over-steers on loose regolith surface.
46. **[Cryptic Error Code Display]**: LTV control screen displays hexadecimal fault code instead of plain-language description -> Crew wastes critical EVA time consulting manual.
47. **[Touchscreen Inoperability in Pressurized Gloves]**: Pressurized glove fingertips fail capacitive touch layer on auxiliary control panel -> Crew unable to toggle camera filters.
48. **[Ambiguous Status Iconography]**: Battery health icon uses non-standard color gradient (yellow for nominal, green for warning), confusing operator -> Crew initiates return too late.
49. **[Audio Prompt Masking]**: Suit ventilation fan noise masks low-volume synthetic voice warnings from vehicle navigation system -> Crew misses off-nominal slope deviation warning.
50. **[Delayed Telemetry Refresh Rate]**: UI display suffers 3-second lag during high data throughput periods -> Operator makes reactive steering adjustments based on past position.
---
### Category 6: Environmental Degradation & External Stressors (51-60)
51. **[Thermal Shock Brittle Fracture]**: Transitioning rapidly from 250 sunlit exterior to 40 K PSR shadow causes structural bracket on sample box to snap -> Box detaches from rack.
52. **[Electrostatic Regolith Deposition]**: High-energy solar wind charges abrasive sub-micron dust, causing it to cling thickly to stereo camera lenses -> Complete optical obscuration.
53. **[Hard Vacuum Cold-Welding]**: Unlubricated titanium bolt threads on drill mount cold-weld together in vacuum -> Impossible to dismount jammed drill payload manually.
54. **[Extreme Low-Angle Glare Blindness]**: 1-degree solar elevation casts blinding horizontal rays that penetrate multi-layer visor coating -> Crew completely loses visibility of near-field terrain.
55. **[Crater Wall Scree Micro-Avalanche]**: LTV wheel vibrations trigger localized downward flow of loose regolith scree -> Vehicle buried up to wheel hubs on crater slope.
56. **[Cosmic Ray Bit Flip in Controller]**: Galactic Cosmic Ray (GCR) strike corrupts memory register in motor control unit -> Controller executes unauthorized reverse wheel command.
57. **[Micrometeorite Puncture of Tire Mesh]**: High-velocity micrometeoroid strikes Nitinol compliant tire mesh, leaving jagged tear -> Progressive structural unravelling during drive.
58. **[Sublimation Vapor Cloud Obstruction]**: Escaping volatiles from core sample sublime violently upon opening canister in vacuum, forming local gas cloud -> Impedes optical sensors temporarily.
59. **[Regolith Abrasion of Seal Surfaces]**: Airborne abrasive dust settles on cryogenic seal flange before closure -> Microscopic gap prevents vacuum seal of sample box.
60. **[Deep Cryogenic Brittleness of Cabling]**: External sensor wiring harness routed along chassis loses all jacket flexibility at 40 K -> Bending over crater rim snaps internal copper conductors.
---
### Category 7: Consumables, Utilities & Auxiliary Dependencies (61-70)
61. **[PLSS Cooling Loop Capacity Depletion]**: High physical exertion during manual core drilling exhausts PLSS sublimator ice reserve prematurely -> Suit internal temperature spikes.
62. **[Auxiliary Battery Bus Voltage Sag]**: Heavy simultaneous draw from drill, headlights, and active thermal control causes auxiliary DC bus voltage sag -> Navigation computer reboots.
63. **[Primary Power Cabling Thermal Overload]**: High-resistance connection at main battery terminal overheats under sustained load, melting insulation -> Partial power loss to drive system.
64. **[Suit Oxygen Supply Line Kink]**: Twisting xEVAS body to position drill causes oxygen umbilical to catch and kink behind life support backpack frame -> Restricted gas flow.
65. **[Hydraulic Fluid Viscosity Spike]**: Actuator hydraulic fluid reaches 40 K limit, increasing viscosity past pump design threshold -> Drill deployment arm freezes mid-swing.
66. **[LTV Battery State-of-Charge Miscalculation]**: Low temperature reduces effective battery capacity faster than predictive algorithm estimates -> Remaining operational time drops by 40%.
67. **[Water Ice Accumulation on Sublimator]**: Excessive moisture freezes across PLSS exhaust port inside PSR -> Suit cooling efficiency drops rapidly, inducing thermal strain.
68. **[Instrument Power Bus Fuse Blown]**: Surge from lighting ballast failure blows main instrumentation fuse -> Loss of all telemetry display panels inside vehicle cab.
69. **[Regolith Contamination of Umbilical Quick-Disconnect]**: Dust enters suit service port during auxiliary power hookup -> Failure to achieve positive seal on auxiliary cooling line.
70. **[Consumable Reserve Margin Ingestion Error]**: Crew relies on nominal consumption tables, failing to factor in 15% terrain grade penalty -> Return trip threatens absolute contingency reserves.
---
### Category 8: Cross-Domain Coupling & Cascading Failures (71-80)
71. **[Dust-Induced Thermal Trap]**: Electrostatic regolith coats radiator panels -> Radiators cannot shed heat -> Motor controllers overheat -> Inverters shut down -> LTV strands crew in PSR.
72. **[Vibration-Induced Loose Connector to Driver Error]**: LTV chassis vibration loosens steering feedback sensor plug -> Steering feels vague -> Operator over-corrects -> LTV slides off designated path into boulder field.
73. **[Cold-Induced Sensor Error to False Abort]**: 40 K cold-soak causes temperature sensor in sample box to read false high -> Control system aborts drilling -> Crew misdiagnoses as sample contamination and panics.
74. **[Lighting Failure to Spatial Disorientation]**: LED headlight failure inside PSR -> Crew activates helmet lights -> Glare reflects off floating dust kicked up by boots -> Crew experiences disorientation and trips.
75. **[Tire Mesh Damage to Control Instability]**: Nitinol tire mesh failure reduces traction on one corner -> Autonomous drive system applies differential braking to compensate -> Overheats remaining motor controllers -> Complete drive failure.
76. **[Drill Jam to Physical Strain]**: Drill auger seizes in permafrost -> Operator applies frantic manual upward force in pressurized suit -> Tears xEVAS glove outer restraint layer against sharp drill casing.
77. **[Comm Blackout to Navigation Error]**: Loss of relay comms inside crater prevents real-time path updates -> Crew relies on degraded dead reckoning -> Enters impassable crevasse field.
78. **[Battery Sag to Reboot Loop]**: Voltage sag from motor surge causes navigation computer to reset -> Computer loses waypoint memory -> Crew disoriented in featureless crater floor.
79. **[Thermal Blanket Loss to Mechanical Freeze]**: Insulation blanket tears off on crater rock -> Direct radiative cooling freezes steering actuator grease -> Steering locks during steep climb.
80. **[Suit Fan Noise to Missed Warning]**: High fan speed noise in helmet masks critical vehicle warning tone -> Crew ignores expanding battery fire warning until smoke enters cabin enclosure.
---
### Category 9: Edge Cases, Boundary Transitions & Latent Conditions (81-90)
81. **[Uncharted Thermal Contraction Crack]**: LTV crosses invisible, freshly formed thermal crack in PSR floor not mapped by orbital reconnaissance -> Wheel drops completely, wedging chassis.
82. **[Dormant Software Race Condition]**: Concurrent LiDAR point-cloud processing and mast pan command trigger rare multi-threading race condition -> Navigation display freezes indefinitely.
83. **[Boundary Transition Traction Drop]**: Transitioning from sunlit high-density regolith to ultra-low-density fluffy PSR permafrost causes sudden wheel spin-out and sinkage.
84. **[Latent Firmware Bug in Drill Torque Limit]**: Unverified edge-case firmware routine fails to scale torque limit at temperatures below 50 K -> Twist-off of entire drill string on first rotation.
85. **[Solar Incidence Angle Calculation Edge Case]**: Mission timing pushes EVA past exact window where long shadows vanish, plunging solar angles into disorienting zero-contrast band.
86. **[Cryogenic Volatile Outgassing Boundary Limit]**: Core sample contains unexpected high-pressure clathrate hydrates that exceed design burst pressure of secondary containment seal.
87. **[Standoff Distance Algorithm Boundary Failure]**: Autonomous obstacle avoidance algorithm encounters vertical ice wall inside PSR; radar cross-section is near-zero, causing LTV to collide directly.
88. **[Latent Battery Cell Manufacturing Defect]**: Deep cold soak triggers internal short in single dormant pouch cell within LTV battery array -> Thermal propagation across module.
89. **[Suit Pressure Regulator Hysteresis Edge Case]**: Rapid descent down 15-degree rim causes barometric pressure sensor in suit regulator to lag, inducing momentary ear block and vertigo.
90. **[Registry Overflow in Telemetry Packet Buffer]**: 8-hour continuous sortie exhausts telemetry logging buffer wrap-around allocation -> Ground control loses last 30 minutes of historical diagnostic logs.
---
### Category 10: Recovery & Secondary Response Failures (91-100)
91. **[Counterproductive Manual Override]**: Operator attempts to force-steer LTV out of regolith drift manually while autonomous traction control is active -> Fighting the system causes complete loss of directional control.
92. **[Emergency Abort Sequence Timing Error]**: Crew triggers abort-to-return sequence before securing sample box lid -> Cryogenic volatiles vent entirely during high-speed return transit.
93. **[Improper Winch Attachment Point]**: During attempted vehicle recovery, crew attaches tow cable to fragile sensor mast base instead of structural chassis towing eye -> Mast shears off.
94. **[Faulty Manual Brake Release]**: Crew attempts to disengage seized wheel brake manually using improper tool on xEVAS tether -> Strips release bolt, permanently immobilizing wheel.
95. **[Over-Correction on Steep Incline Escape]**: When LTV loses traction climbing out of crater, driver over-corrects steering uphill -> Vehicle rolls laterally down the 15-degree slope.
96. **[Incorrect System Reboot Sequencing]**: After computer lockup, crew power-cycles main vehicle bus before saving diagnostic state -> Corrupts persistent navigation map data.
97. **[Emergency Suit-to-Suit Transfer Mishap]**: During emergency contingency oxygen sharing, crew members entangle umbilical hoses in LTV handrail -> Restricts movement and damages quick-disconnects.
98. **[Premature Abandonment of Vehicle]**: Crew prematurely initiates vehicle walk-back protocol due to false-positive battery fire alarm, exposing themselves to excessive walking fatigue 12 km from lander.
99. **[Faulty Manual Override of Drill Extraction]**: Operator uses raw suit torque to pull jammed drill without reversing auger rotation -> Bends drill string inside borehole, trapping drill permanently.
100. **[Incorrect Emergency Beacon Frequency Selection]**: In panic following comms blackout, crew selects backup beacon frequency offset from lander receiver band -> Lander search team fails to locate stranded crew.
From <Anaconda 3 Jupyter Notebook "What_Can_Go_Wrong.ipynb" run>
Comments
Post a Comment