Manufacturing Technology Project Turns Metal 3D Printing Into a Predictive Process

A new American startup is aiming to transform metal additive manufacturing from a trial-and-error process into a predictive, physics-based workflow. Manufacturing Technology Project (MTP) has launched its flagship platform AdditiveM, which combines artificial intelligence, multiphysics simulation, and digital twins to optimize metal 3D printing processes.

Bridging the Gap Between Simulation and Reality

Founded by Hamed Hosseinzadeh, a computational mechanical and materials engineer, MTP aims to bridge the gap between high-fidelity physics simulation and real-world manufacturing practice.

"What immediately fascinated me was that additive manufacturing is not just a fabrication method, but a fully coupled physical system in which heat transfer, fluid flow, phase transformation, residual stress, microstructure evolution, and structural performance are all tightly interconnected," Hosseinzadeh explained. "At the same time, I observed that much of the industry still relied heavily on trial-and-error and empirical tuning, which is costly, time-consuming, and risky for qualification-critical applications."

AdditiveM: A Digital Twin for Metal 3D Printing

AdditiveM is designed to act as a digital twin of the printing process, capturing the full thermomechanical history of a part and linking it to residual stress, distortion, microstructure, and eventually fatigue and performance.

The platform's core features include:

• High-fidelity transient thermal simulation of layer-by-layer melting and heat accumulation
• Thermo-elasto-plastic mechanical modeling with temperature-dependent plasticity, strain hardening, cyclic hardening, and Bauschinger effect
• Prediction of residual stress, distortion, and crack-prone zones
• Fatigue-relevant stress-strain history and performance indicators
• Stochastic and AI-assisted modules for anomaly-aware and uncertainty-aware prediction
• Direct link between G-code, scan strategy, material properties, and machine parameters and final structural integrity

Process-Agnostic Approach

AdditiveM has been developed as a process-agnostic computational framework rather than a single-process tool. Currently, the platform focuses on laser-based Powder Bed Fusion (PBF) and Directed Energy Deposition (DED), with plans to extend to electron beam and arc-based systems.

The roadmap includes compatibility with multi-laser and multi-beam platforms, with full consideration of scan dynamics such as velocity, acceleration, and jerk—which strongly influence melt pool stability, thermal gradients, and defect formation.

Four Key Challenges Addressed

From an industrial perspective, AdditiveM addresses four fundamental challenges:

1. Qualification-grade predictive tools — Many failures in metal AM originate from residual stress, elasto-plastic deformation, and microstructurally driven fatigue. AdditiveM addresses this through coupled thermo-elasto-plastic modeling.
2. Process-to-performance connection — The platform establishes a true link between scan strategy, laser power, speed, layer thickness, and material properties and the final part performance.
3. Physics-based database — Creates validated physics-based data for AI-assisted learning.
4. Computational accessibility — Highly optimized numerical algorithms and reduced-order solvers run on standard engineering workstations and even high-end laptops.

Autonomous Production Control

The ultimate goal? Autonomous production control. AdditiveM is being developed to support design, process optimization, qualification, and ultimately closed-loop, intelligent control of metal additive manufacturing systems.

"In this sense, AdditiveM will not be just a simulator; it will evolve into a Physics-Based Digital Twin platform enhanced by AI," Hosseinzadeh stated.