40-60% reduced rejection in metal manufacturing processes

Detect quality deviations before they occur and use AI recommendations to reduce scrap

Download Case Study

The Change we bring
TVARIT Industrial AI for Die Casting

Up to 1.500.000 € of savings per plant

by reducing waste of man, machine, material & energy

Up to 5% increased OEE

through increased productivity & reduced scrap

Up to 60% reduced scrap

with suggestions for optimal casting parameters

up to 7,2 GJ reduced energy consumption per plant

due to reduced rework

Find out how a

A world leading wheel manufacturer reduces scrap in LPDC with TiA

Download Case Study

Testimonials

"Working with Tvarit is incredibly refreshing. Their young team of smart talents helps us to successfully implement the digital transformation of our company. The particularly high level of commitment and dedication that Tvarit brings to the joint projects is impressive. I am pleased about the positive results achieved so far and those to come.“

Dr. Alireza Tavakoli

Vice President

Digitalization/CDO

"As CIO of a global company, I have gained a lot of experience with solution providers. What makes Tvarit stand out here is their deep understanding of the manufacturing processes and their challenges. With a deep insight in focus areas to address for improvements, Tvarit has been able to deliver a high degree of innovativeness within short project implementation time. Tvarit industrial AI (TiA) provides clearly understandable insights into the manufacturing processes and helps to achieve objective and measurable results (such as improving OEE and the reduction of scrap).”

Esteban Remecz

Vice President & CIO

The way it works

capture your knowledge from your data with process specific artificial intelligence

Process specific hybrid AI models are used to analyse how the various influencing factors in the casting process affect the quality of the casting.

Quality predictions help to recognize deviations in real-time

Root cause analyses define the reasons for failures

Recommendations for optimal casting settings help to reduce scrap

Schedule Demo

See deviations at a glance

Monitor your manufacturing processes in real-time

Detect quality deviations before they occur with the TiA product quality prediction

Receive notifications via e-mail, SMS, teams to integrate into existing workflows & share insights with team members

Understand & act
to reduce rejections

Understand the reasons for deviations with scrap root-cause analyses

Reduce scrap with suggestions for optimal casting parameters

Trustworthy hybrid model - completing the puzzle with metallurgical knowledge

"AI kisses metal"

TiA combines metallurgical knowledge (such as phase transfer equations & metal flow simulations) with machine learning for more stable, accurate & reliable results.

Get a deep insight into our Software TiA for diecasting through this success story demonstrated by our Head of Die-casting divison, Tobias Gundermann.

Would you like to know more ?

See TiA in action & get answers on your questions from our experts

Schedule demo

The team behind TiA for metal processes

Tvarit GmbH is a Frankfurt-based software company specializing in the application of artificial intelligence to increase efficiency in manufacturing processes.

Our team consists of over 30 highly qualified engineers and experienced data scientists. We combine extensive industry experience with academic research and development - driven by a passion for technology & innovation.

Our customers include world-renowned manufacturing companies in the metal industry such as Aditya Birla, Kamax and Maxion Wheels. What they particularly appreciate about working with us is our high level of flexibility and personal support, our high degree of innovativeness and our fast project implementation times.

In addition to proven results from more than 45 industrial projects in reducing scrap, increasing machine availability and reducing CO2 emissions, we have been recognized by the EU as one of the best 8 out of 490 AI companies across Europe. (EDI, 2020)