UC- Product Change Impact AI Estimator

Description

This use case outlines how the system evaluates the impact of a proposed product or component change across the supply chain, inventory, and compliance domains. By leveraging AI models trained on historical change management data, the system automatically classifies the impact as Low, Medium, or High risk. It then generates a structured recommendation report identifying the required documentation, stakeholder approvals, and actions needed prior to product rollout.

Actors

• Product Manager – Initiates the product/component change request.

• Engineering Lead – Reviews technical impact details.

• Compliance Officer – Validates compliance obligations.

• Supply Chain Manager – Evaluates impact on logistics and vendors.

• AI Estimator System (Power Platform) – Performs automated impact analysis and report generation.

Preconditions

• The change request must be formally submitted through Power Apps with all mandatory fields (product ID, component details, proposed changes, justification, etc.).

• Historical change data must be available in Dataverse for AI training and inference.

• AI model is trained and published in Azure AI Builder.

• Necessary Power Automate workflows are in place to trigger the AI estimation and notification process.

Flow of Events

Main Flow

1. Product Manager submits a product change request via Power Apps.

2. The request data is saved to Dataverse.

1. Power Automate triggers the AI model built in Azure AI Builder to analyze the impact using historical patterns.

1. The AI model classifies the change as Low, Medium, or High risk.

2. Based on the risk level:

   • A recommendation report is generated including suggested documentation, compliance checkpoints, and stakeholders to involve.

   • Notifications are sent to relevant actors.

3. Product Manager and relevant stakeholders review the report.

4. If accepted, the change process proceeds to formal approval and rollout planning.

Alternate Flow

• If the AI model returns insufficient confidence or flags missing data, the request is redirected for manual review.

• The Product Manager is notified to provide additional information.

Postconditions

• The change request is classified and documented with a corresponding AI-generated impact report.

• Stakeholders are notified and aligned based on the risk level.

• The product change process is either escalated, approved, or paused based on AI recommendations and human review.

Benefits

• Reduces manual effort in evaluating complex change impacts.

• Improves decision-making speed and accuracy using AI-driven insights.

• Ensures better compliance by auto-recommending documentation and stakeholder reviews.

• Standardizes the evaluation process across teams.

Tools & Technology Used

• Power Apps – Frontend interface for submitting change requests.

• Dataverse – Central data repository for change request metadata and historical logs.

• Azure AI Builder – Trained model that evaluates impact and classifies risk.

• Power Automate – Orchestration of workflows between components (data input, AI prediction, report generation, notification).