Umetrics Suite Blog

Injection molding no longer a black art: How to increase volume while reducing defects

February 26, 2018

For manufacturing companies, process control is essential— even for those producing low-cost items such as small plastic parts. That’s because even when units are small and inexpensive, the cost of defects becomes exponentially higher when they reach the next manufacturing step at another plant.


Injection molding is no longer a black art if you have the right data to inform your processes.


This is especially true in the pharmaceutical and medical device industries, as well as in other industries with rigorous regulatory control. So what can you do to root out the small problems before they become bigger ones?

We can see how this plays out by looking at an example from a pharmaceutical company with a medical device division that uses injection molding to produce small plastic components.

Step 1: Realize traditional quality control is insufficient

With injection molding, process defect rates are usually very low. The processes are very stable and a particular defect may only last two or three cycles out of perhaps a hundred thousand cycles a day.

But does that mean there are no problems? We know that faults exist, even minor ones, and that finding them can help save money and reduce problems down the road. So what can we do?

To detect a component default in this situation becomes almost impossible using traditional quality control methods. You simply must have real-time data and multivariate methods to solve the problem.

Step 2: Ensure your data is accurate

That’s exactly what our example company decided. They embarked on a project to implement real-time data release on their injection molding process. How? The first thing you have to do is to ensure that you have good data. That often means starting very old school by applying analog and digital sensors, and starting to understand your process and your machines.

Step 3: Monitor your data in real-time

Once you have ensured you have quality data, you can continue to the next step – monitoring the data in real-time. Using software such as SIMCA and SIMCA-Q, you can create a multivariate ideal model of the process, like our example company did, which can then be run in real-time to detect deviations from a normal production cycle.

Step 4: Use actionable analytical and visual tools

To make everything work, another piece of the puzzle is to have actionable analytical and visual tools. The operators need easy-to-understand insights to adjust the process and to ensure that the machines are producing quality products.

On the management level, our example company used Active Dashboard, where data from thousands of machines is aggregated in one single dashboard. It gives a production overview and provides tools to drill down and focus on a single plant and even on a single machine.

Step 5: Measure your results

What sort of results can you expect?

Starting with just one machine, our example company was able to increase production considerably without any additional headcount. They also saw a more than 90 percent reduction in the number of non-conformances. Ramping that up to the bigger picture, with thousands of machines running in plants 24/7 and having real-time data available 100 percent of the time, translates into a fundamental business impact.

What sort of impact could real-time data monitoring have on your processes?

Injection molding no longer a “black art”

In this particular industry, injection molding has been considered a “black art”. You need years of training to understand the process and even after that you have to rely on intuition. What this example shows is that you can gain a tremendous amount of understanding – and save costs – by implementing real-time monitoring of your production.

Watch this video to find out more about real-time data monitoring

Find out more

Find out how SIMCA and Active Dashboard can help you get a better view of your production processes and help you make better decisions for your business.  Get our presentation and access an online demo.

Get the presentation







Topics: Statistical Process Control, Manufacturing Quality Control, Manufacturing Processes

Jonas Elfving

Written by Jonas Elfving

Product Manager at Sartorius Stedim Data Analytics