AI grows in importance to injection molders, OEMs say

Nov. 16, 2023
Arburg, LS Mtron and Milacron are devising ways to use the technology to save time, money, energy and resin.

By Bruce Geiselman 

The incorporation of artificial intelligence (AI) to boost efficiency and reduce energy and material consumption holds growing interest for buyers of injection molding machines, according to some of the largest equipment manufacturers. 

“AI is becoming increasingly important because of the need to automate injection molding processes efficiently and flexibly despite ever-smaller batch sizes and shorter product life cycles,” said Werner Faulhaber, VP of research and development at Arburg. 

“Application examples of AI include automatic programming of robotic systems, targeted malfunction remedying, and a spare-parts system with ‘intelligent’ image processing. Arburg is working step-by-step to make injection molding ‘more intelligent,’ ensuring that the machine continuously learns, keeps itself stable, and can even optimize itself in the future.” 

AI can be particularly important when setting up the initial molding process for a new molded part, as there can be a deviation in production efficiency depending on the proficiency of the process engineer, said Peter Gardner, president of LS Mtron Injection Molding Machine USA. 

LS Mtron has been developing its own AI-based smart injection molding system primarily with affiliated companies and suppliers in South Korea, but hopes to launch the offering to U.S. customers next year. It will demonstrate its Smart Solution 4.0 from May 6-10 at NPE 2024 in Orlando, Fla 

“It is new, and we’re not really even offering it in the U.S. right now,” Gardner said. “We are going to roll this out at the NPE with some demonstrations in booth in some machines that are doing various AI functions … We will be doing demonstrations at the show, and then, hopefully, commercially offering it to our customers.” 

AI technology can imitate the experience of highly skilled molding process experts to recommend optimal process settings to reduce the initial process setup time and the risk of becoming too dependent on the availability of highly skilled workers. AI allows lessons learned from existing company plants to automatically be transferred to new plants without the need to send personnel to the new location. 

LS Mtron’s AI Process Setup Assistant system, part of the Smart Solution 4.0 package, has reduced the initial process stabilization time by 23 percent on average among customers, according to Gardner. 

Milacron also is evaluating AI and augmented reality for multiple applications across its plastics processing equipment from customer products to training to business performance improvement. 

Milacron plans to unveil some innovative solutions in May at NPE, said Bunlim Ly, director of strategic marketing and innovation at Milacron.  

A team at the company’s global headquarters near Cincinnati is using AI and machine learning to optimize process controls and real-time digital twin technology to shorten the design and validation process, according to the company.  

“In addition, they have been working to generate IoT [Internet of Things] data to enhance machine maintenance,” according to a Milacron statement. 

Details on Arburg AI 

Working with simulation partner Simcon, Arburg in 2021 introduced its aXw Control FillAssist technology to Allrounder injection molding machines equipped with the Gestica machine control system 

The Gestica control automatically knows which material it is processing and which component it is producing. The aXw Control FillAssist supports the mold setup process, finding an initial ideal operating point. In addition, it offers a display on the control that simulates the filling level of the component. 

“During this active setup, the user can see into the cavity as if they were wearing ‘virtual glasses’ and observe the component filling process live,” Faulhaber said. 

With the visual guidance, it is easy to find the right fill-to-hold transition point and ensure that the part can be demolded without any mold damage. The time and costs required for setup can be reduced by more than 40 percent with this filling simulation, Faulhaber said.  

“It is important to mention that human operators retain the decision-making power at all times — not the AI system,” he said. 

More recently, Simcon expanded the simulation capabilities of the FillAssist technology with a new plug-in called Varimos that uses AI to predict the effect of a parameter variation before a component is even produced. 

“Varimos automates the repetitive portions of this process,” Faulhaber said. “It makes it possible to quickly create many variants of the injection molded part, mold and parameters, simulate them in parallel and analyze the results using artificial intelligence.” 

The design of the molded part and mold can be shortened from weeks to a few days, according to the company. 

In the future, Arburg hopes that through the use of digital twin technology and simulating production cycles, it will be possible to make energy predictions in advance. 

LS Mtron Smart Solution 4.0 

LS Mtron’s new Smart Solution 4.0 package includes the Setup Assistant and the AI Weight Control systems.  

The Setup Assistant finds the optimal initial injection molding parameters of injection molded parts through an AI algorithm that extracts key geometrical information from molded part modeling, derives the best molding process parameters and applies them to the injection molding machine, Gardner said. AI determines factors including the appropriate mold temperature, injection size, injection velocity, pack time and cooling time. 

The AI Weight Control System works either with an external scale to weigh parts or a weighing system or loadcell that is mounted on a take-out robot. It ensures that part weight remains consistent. When a deviation occurs, the AI engine automatically and immediately applies new molding process parameters to compensate for the deviation.  

Issues that can cause weight fluctuations can include differences between batches of raw material or recyclate, and fluctuations in the process environment, including factors like ambient temperature and humidity. 

Users of the LS Mtron AI technology can decide whether they want to keep all their processing information stored exclusively on a local server or if they want to share some of their information anonymously on the cloud and benefit from knowledge the AI software obtains from other users. 

“The end user can choose how deeply they want to use this, or do they just want to use it to help set up their molds easier?” Gardner said. “Do they want real-time adjustments to be going on based on the actual weight of the parts that are coming out of the machine and based on thousands of other molding machines that are running, perhaps similar products or similar types of processes?” 


Arburg Inc., Rocky Hill, Conn., 860-667-6500,  

LS Injection Molding Machine USA, Duluth, Ga., 470-724-2263,  

Milacron LLC, Batavia, Ohio, 513-536-2000,  

About the Author

Bruce Geiselman | Senior Staff Reporter

Senior Staff Reporter Bruce Geiselman covers extrusion, blow molding, additive manufacturing, automation and end markets including automotive and packaging. He also writes features, including In Other Words and Problem Solved, for Plastics Machinery & Manufacturing, Plastics Recycling and The Journal of Blow Molding. He has extensive experience in daily and magazine journalism.