2D Descriptive Melt Pour Process Digital Twin

MxD and XMPro are implementing a 2D descriptive digital twin at a New Jersey manufacturer powered by advanced digital engineering methodologies that integrate existing process data with new sensor inputs to optimize melt pour processes, capture institutional knowledge, and improve product quality by transforming subjective expert judgment into standardized, data-driven operations.

Project Awarded:

July 15, 2025

Project Number:

24-12-02

Project Lead

XMPro

Problem

Manufacturing efficiency and quality are significantly compromised by outdated operational methods. The reliance on multiple manual processes introduces inconsistency, increases error rates, and prevents optimal production flows. Product quality assessment depends entirely on the subjective judgment of veteran operators with years of specialized experience, creating a critical knowledge gap as these experts retire. The current training approach relies exclusively on time-consuming hands-on apprenticeship, resulting in extended onboarding periods and inconsistent skill development. Without systems to capture, standardize, and transfer institutional knowledge, the organization faces escalating risks to production continuity, quality standards, and competitive position as workforce demographics shift.

Proposed Solution

The team will implement a 2D descriptive digital twin to help optimize and improve the quality of the pours. This solution will aim to improve quality of the product utilizing new and existing data to optimize pour times, integrate existing process data with new sensor data for valuable insights, provide the ability to record manual information to assist with “deep dive” investigations in the future, and create a library of “material recipes” for operations and engineers capture and adjust future batch data.

Impact

The 2D descriptive digital twin for melt pour processes will impact manufacturing in several practical ways such as, preserving critical knowledge from retiring experts by converting their experience into digital records and decision models, reducing training time for new operators who can learn from standardized processes rather than years of apprenticeship, improving product consistency by replacing subjective quality assessments with data-driven parameters and, decreasing material waste through more precise control of pour conditions The core impact is transforming manufacturing from an art dependent on few experts to a science that can be taught, measured, and systematically improved.