BlinqIO Documentation

AI Recovery Case Study On OrangeHRM ​

This case study outlines how the AI Recovery feature enables test automation to self-heal when faced with UI changes demonstrated through an OrangeHRM example.

Existing Test ​

A test script automates the process of logging into OrangeHRM, adding a new employee, verifying leave balance, and logging out using recorded locators and stable UI elements.

orangehrm-1

Test Fails Due to UI Change ​

An update to the OrangeHRM interface causes locator failures. Elements move, layouts change, and the test begins to fail at runtime due to mismatched selectors.

image-20250711171346199

Execution Flow with AI Recovery ​

When you run the test with AI Recovery enabled either manually from the UI or automatically through your CI/CD pipeline, the system goes through the following phases:

image-20250711171536055

Note

Refer Execute scenario with AI Recovery to learn more.

1. Initial Run – Fails at Outdated Locator ​

The test executes using previously recorded logic and fails when an element is not found.

image-20250711171744453

image-20250711171835643

This is how the Reports page displays the failed step, with the error message, root cause, and retraining reason highlighted.

image-20250714172634579

2. Root Cause Analysis – AI Learns & Regenerates ​

AI Test Engineer identifies failure, inspects the updated DOM, finds potential new matches based on context and labels, and dynamically regenerates locator logic.

image-20250711171934312

image-20250711173113539

Note

Refer Root Cause Analysis to learn more.

3. Final Run – Test Succeeds with Regenerated Steps ​

The scenario re-executes with the new locator. It successfully interacts with the updated element, completes execution, and marks the recovered step with an AI Recovery badge.

image-20250711173149748

The recovered step appears with a green status and an Error fixed by AI badge, indicating successful self-healing.

image-20250714173416369

This OrangeHRM case study highlights how AI Recovery keeps your test automation resilient in real-world conditions. It instantly responds to UI changes by repairing broken selectors and retrying the failed step, all without manual effort.

By eliminating the need for re-recording and reducing maintenance overhead, AI Recovery helps teams maintain test stability and keep CI/CD pipelines running smoothly even as applications evolve.