Industry Insights

"CHINAPLAS 2026 in Shanghai came to an end this week. I wasn't able to attend the event in person. But these past few days, I've been reading various post-event reports, technical news, and friends' posts from colleagues. Although I didn't see those new machines with my own eyes, several changes have indeed made me feel that the gameplay in the next two or three years might be different."

📡https://www.chinaplasonline.com/CPS/event/Smart-Molding-Shaping-the-Future/simp

Let me talk about the change that has touched me the most: AI really entered the production line, and it's not just a gimmick.

Over the past two years, terms like "AI Injection Molding" and "Intelligent Factory" have been frequently mentioned. But I have always been somewhat skeptical.

On one hand, our industry is based on experience. How to use the right material at the right temperature, how much pressure to apply, how to divide the speed into segments - many of these things were learned by veteran workers over many years. Before AI was fully developed in the previous two years, I wasn't too convinced that it could have such strong learning capabilities.

On the other hand, the so-called "intelligence" I had seen before mostly involved installing a large screen on the machine to present the data beautifully. But when it comes to adjusting the machine and solving problems, it still depends on human intervention.

So before CHINAPLAS, my attitude towards AI Injection Molding was basically: It sounds good, but it's still far away from me.

After reading the news about this week's exhibitions, my thoughts changed.

The "iChe Smart Family" intelligent injection molding ecosystem released by Zhenxiong has already enabled a one-click start-up solution for intelligent factories, and it also has an AI injection molding assistant - you can directly ask it how to adjust the process parameters, and it can give suggestions.

It is reported that this system can analyze over 1,000 process parameters in real time, detect deviations early, and automatically adjust and optimize the process. What does 1,000 parameters mean? An experienced master actually only focuses on 20 to 30 key parameters. Machines can simultaneously view 1,000 parameters and are constantly observing - the gap in this dimension cannot be made up by experience.

Lijin's injection molding machines have also taken action. The control accuracy of the fully electric medical-specific machines can reach 0.01mm, and the weight deviation is ≤ 0.1%. To be honest, for us who produce precision parts, sometimes the scraps occur because of that zero point zero several millimeters' fluctuation. If we can stabilize at this precision, the space for improving yield is very large. Moreover, for medical equipment, this intelligent change may improve the cure rate of patients on the instrument, which is a very promising idea.

For factories, as technology progresses, people can only follow. This transformation might make some employees uncomfortable. When the hand-crank machine was replaced by the computer machine, there was also a group of people who were not adapted. But people should keep up with the times and keep up with the pace of development to avoid falling behind.

One concern I have is that if AI is to be fully applied to production, it may require very high costs. But on second thought, there might not be no chance at all. Because we don't need so many basic employees, this can also achieve a balance, and AI always has its own program, with a greatly reduced error rate.

Our company's outlook:

In the short term, we aim to standardize the existing data in the workshop (without data, even the most advanced AI is of no use); have the young technicians gain more experience and knowledge through practice; and pay attention to cost-effective intelligent solutions. Finally

My impression of CHINAPLAS 2026 is this: The technological turning point in the injection molding industry might truly have arrived.

Over the past decade, we have been making progress in materials, molds, and equipment, but it has been linear and gradual. However, the entry of AI into the workshop might bring about a non-linear leap - just like how smartphones replaced feature phones, not just improving a little bit, but a completely different way of doing things.

To be honest, I am also a little anxious. After all these years of work, many of our experiences might become less valuable in the face of a set of algorithms. But from another perspective, if our industry always relies on senior workers passing on knowledge and skills to juniors, and the scale never expands, and efficiency never improves - then that would be even more worrying.

The arrival of AI is both a challenge and an opportunity.