> ## Documentation Index
> Fetch the complete documentation index at: https://guide.likemind.asia/llms.txt
> Use this file to discover all available pages before exploring further.

# 7. Running an Investigation

Open a case and switch to the Investigation workspace — a three-pane layout with evidence on the left, the Logic Canvas in the middle, and CAPA actions on the right.

## 7.1 5 Whys

Linear cause chain — ask Why up to 5 times. AI suggestions are capped at 7 levels deep.

## 7.2 Fishbone (Ishikawa)

Groups causes into the 6M categories: Man, Machine, Method, Material, Measurement, Environment. Default skeleton renders the effect, 6 categories and 12 causes; add more by hand, up to 2 levels deep per branch.

## 7.3 Fault Tree Analysis (FTA)

Top-down logic tree with AND / OR gates and basic / intermediate events.

## 7.4 Swiss Cheese Model

Five standard barriers: Organizational, Procedural, Engineering, Supervision, Unsafe Acts. Holes represent failed defenses; alignment shows accident pathways.

## 7.5 Root cause classification

Each cause node can be classified as Direct, Indirect or Contributing — these classifications carry into the AI summary and PDF report with a 1:1 mapping.

## 7.6 AI help during investigation

* AI Cause — suggests next-level causes based on the current node
* AI CAPA — drafts corrective and preventive actions
* AI Summary — generates a category-aware executive summary
* Right-click any node to edit, delete or link evidence