MARVIN tools
Module 03

Probabilistic Margin Analysis

Map components onto a Design Structure Matrix, capture the likelihood and impact of change propagation between them, and find the couplings most likely to consume your margins.

Overview Organisation Tutorial Case Studies Literature
01

Overview

Margins do not fail in isolation. A change introduced anywhere in a system ? a new requirement, a different supplier, or a redesigned subassembly ? tends to propagate along the couplings between components.

This module captures couplings as a Design Structure Matrix (DSM) and runs Clarkson's Change Prediction Method (CPM) on it. The result is a matrix of combined risk values showing how likely change is to propagate through the network, and how large the expected impact is.

02

The CPM method

3

The module follows Clarkson's CPM with a margin-focused interpretation of propagation risk.

Step 01
Direct likelihood

For each ordered pair of components, capture the probability that a change in the source will require a change in the target.

Step 02
Direct impact

Capture the average severity of the propagated change: the likely size of the redesign triggered across an interface.

Step 03
Combined risk

Compose direct probabilities along all paths to produce combined likelihood, impact, and risk matrices.

03

What the module supports

Feature
Interactive DSM editor

Add components, group them into subsystems, and edit direct likelihood and impact in a matrix view.

Feature
Combined risk heatmap

Visualise CPM results as a heatmap on the same DSM to spot hotspots and dominant paths.

Feature
Per-element profiles

Inspect incoming and outgoing risk for any component, ranked by likelihood, impact, or combined risk.

Feature
What-if exploration

Edit a direct coupling and see how the combined-risk matrix responds before committing to a redesign.

04

Where to go next