In injection molding, a part can pass inspection and still raise a question. What if its measurements are slowly moving toward the edge of the allowed tolerance? What if one production run looks different from the last, even though both technically meet the drawing? A single pass-or-fail result cannot tell the whole story.
Quality assurance needs a view of what is happening across production, not just at one inspection point. That is the idea behind statistical process control, or SPC: collect measurements over time, look for variation and patterns, and use what the data reveals to guide decisions.
At Nylacarb, we have taken that principle a step further by developing a custom application that turns our quality data into SPC charts and helps us monitor how measurements compare with specified tolerances. We used AI and API software integration as part of the application development process, building a tool around the way our team actually works. The purpose is practical: make important changes easier to see and give our people better information for investigating them.
Why a Passing Measurement Is Only Part of the Picture
Every manufactured part has requirements. A drawing may call for a feature to fall within a defined dimensional range. Inspection tells us whether a measured part meets that requirement at the time it is checked.
But imagine a dimension with an acceptable range of 9.90 to 10.10 millimeters. Several samples measure 10.00, 10.02, and 10.04 millimeters. Later samples measure 10.06 and 10.08. Each of those results may still be within specification. Together, however, they prompt a useful question: Is the process shifting?
The numbers in this example are illustrative. The relevant requirements and sampling plan depend on the actual part and customer specifications. The point is that a series of measurements can tell us something a single measurement cannot.
Changes in measurements can have many possible explanations. Material behavior, machine settings, tooling condition, measurement methods, and other production factors may warrant investigation. A chart helps bring the pattern to attention; the quality and production teams still need to determine what it means.
Control Limits and Specification Limits Are Different
One of the most important distinctions in SPC is the difference between control limits and specification limits.
Specification limits come from the part’s requirements. They define the acceptable range for a particular characteristic. A measurement outside that range requires attention under the applicable inspection and quality procedures.
Control limits are calculated from process data using an appropriate statistical method. They help indicate whether the process is behaving consistently with its established pattern. They do not replace the customer’s specifications.
A process can be statistically consistent while consistently producing parts too close to, or even outside, a specification limit. Conversely, a process can produce parts that currently meet specifications while showing a change that deserves investigation. That is why looking only at pass-or-fail results—or treating the two sets of limits as the same thing—can leave an incomplete picture.
For our team, the value of charting is the ability to examine both questions: Does the part meet its requirements, and what is the process doing over time?
What Is Statistical Process Control?
SPC is a method of using process data to understand variation. Rather than waiting for an out-of-specification result, teams examine measurements in sequence and look for signals that the process may be behaving differently than expected.
That distinction matters because variation is a normal part of manufacturing. No two measurements will be perfectly identical. The goal is to understand the variation well enough to recognize when a change deserves attention.
Control charts are one common SPC tool. They plot data over time against limits based on the behavior of the process. Depending on the chart and the data collected, a quality professional can look for points beyond control limits or patterns that suggest a change. The chart provides evidence for an investigation; it does not diagnose the cause by itself.
SPC can help teams answer questions such as:
- Is this process behaving consistently over time?
- Has the average measurement shifted?
- Is the amount of variation changing?
- Do the latest results call for investigation?
- Are the measurements still meeting the customer’s requirements?
Those questions are related, but they are not interchangeable.
A Custom Tool Built Around Our Quality Workflow
Generic software does not always fit the way a manufacturer collects data or reviews a specific part. Nylacarb’s new SPC application was developed to work with our quality data and make charting and tolerance monitoring more accessible to our team.
We used AI to help build the application. That describes how we developed the software, not how quality decisions are made. The system organizes and visualizes measurements; trained people review results, verify concerns, investigate causes, and determine the appropriate response.
This approach reflects a broader belief at Nylacarb: technology is most valuable when it helps people do their work with greater clarity. An effective quality tool should make information easier to use at the point where a decision is needed. It should support the team’s established inspection and documentation practices, not create another isolated set of numbers.
The application can show variation in a form that is easier to interpret than a long list of measurements. It can also highlight when recorded values fall outside defined specifications, helping the team identify results that require follow-up. The chart makes a conversation possible: what changed, when did it change, and what should we examine next?
What This Means for Customers
Customers need more than an assurance that a part looked good at one moment. They need confidence that production is being monitored against their requirements and that the manufacturer has a way to recognize and address concerns.
An SPC-supported quality process can provide several practical benefits:
Better visibility into variation. Reviewing measurements as a sequence helps reveal shifts that may be hard to notice in individual inspection records.
More informed investigations. A chart can show when a change began, giving the team a useful starting point for examining production conditions and measurement data.
Clearer communication. Visual data can make it easier for quality, production, and customer teams to discuss what is happening with a critical characteristic.
A stronger basis for improvement. Over time, measured results can inform adjustments to the process, inspection approach, or other aspects of a project.
No chart can promise a zero-defect process. The benefit comes from pairing reliable measurements with sound interpretation and timely action. It is one part of a larger quality system that includes clear specifications, appropriate inspection methods, documented procedures, and experienced people.
Innovation That Serves the Work
AI is often discussed in sweeping terms. In manufacturing, its most useful applications may be far more specific. We saw a need within our quality workflow and used available technology to create a tool for that need. The result is an application that helps us turn measurement data into a clearer view of production.
For an injection molding partner, innovation should have a purpose. It should help the team ask better questions, spot meaningful changes, and respond with evidence. Our SPC application is an example of that approach: build for a real production challenge, keep qualified people at the center of quality decisions, and continue improving the way work gets done.
If your project requires close attention to dimensional consistency or a more detailed conversation about inspection and quality planning, contact Nylacarb. We would welcome the opportunity to review your requirements and discuss an approach suited to your part.