Co-Pilot Stage 3: Validate Assumptions

Introduction

Welcome to Co-Pilot Stage 3: Validate Assumptions. In this stage, you will review the AI-generated results to ensure they align with your expectations. This step is crucial for confirming that the AI has made appropriate assumptions based on the data you provided. It’s important to remember that the accuracy of these assumptions depends heavily on the quality of your input data.

In this video, you will learn how to:

• Check the Co-Pilot status and review AI-generated results.

• Examine the assumptions made by the AI.

• Make adjustments to input data if necessary.

• Ensure the results align with your expectations before finalising the LCA.

Step-by-Step Guide

1. Checking the Co-Pilot Status:

   • Once the AI has completed the LCA, go to the 'Products' section to see the status of your project. For example, the Trailmaster 5000 should show a Co-Pilot status of "done."

   • You will also receive an email with details about the process duration and the top-level CO2 embodied in the product. Click the link in the email to go directly to the product or navigate through the platform.

2. Viewing and Editing the Canvas:

   • Go to the product and click on "View and Edit Canvas" to start reviewing the AI’s assumptions.

   • The canvas uses colour coding to indicate CO2 levels—red for high emissions and green for lower emissions. Dots beside each element indicate the accuracy of the AI's analysis.

3. Reviewing Resource and Process Nodes:

   • Click on high-emission nodes to review detailed AI assumptions. For example, select a resource node to see the detailed description and decomposition of components.

   • The AI may decompose complex items into subcomponents for a more accurate assessment. Review these decompositions to ensure they are logical and accurate based on your knowledge.

4. Checking Assumptions and Data:

   • Examine the input assumptions taken from your data and how the AI processed them. For example, if a wheel was decomposed into rubber and aluminium parts, check if these assumptions are correct.

   • Scroll through the notes and assumptions to understand the AI’s logic and calculations.

5. Making Adjustments:

   • If you find any discrepancies or areas that need refinement, you can adjust the input data directly in the canvas. For instance, change the weight of a component or update the description to be more accurate.

   • After making changes, re-run the Co-Pilot to update the analysis. Note that this may take a few minutes depending on the complexity of the data.

6. Validating Process Nodes:

   • Similarly, review process nodes for accuracy. For example, check the energy consumption and process steps assumed by the AI for manufacturing components.

   • Ensure the AI’s assumptions about processes (like welding) and their environmental impacts (like electricity and gas use) are accurate.

Next Steps:

Once you have validated and, if necessary, adjusted the assumptions made by the AI, you are ready to proceed to Co-Pilot Stage 4: Rapid Reporting. In the next stage, you will generate comprehensive reports that provide actionable insights for reducing your product’s environmental impact.

• Proceed to Co-Pilot Stage 4: Rapid Reporting

---

This page provides a comprehensive guide to validating the assumptions made by Emvide's AI, ensuring that the lifecycle assessment results are accurate and reliable. Proper validation is key to achieving high-quality LCA outcomes.