Using Augmentation to Reduce Variability in Manufacturing

Simply put, manufacturing variability is the difference between the intended outcome of an action or procedure in your production process, and the actual result. Sources of variability can be difficult to isolate, and they’re just as likely to be attributable to materials, machines, and devices as human performance.

As you can imagine, reducing variability in manufacturing is a crucial step toward maximizing value.

One of the best ways to reduce variability throughout the manufacturing process is augmenting workers with performance-enhancing technologies. These technologies help workers perform more consistently, detect quality issues early, and arm them with the data they need to make informed decisions.

This blog will show how manufacturers can use Industry 4.0 technologies to reduce variability in processes and outcomes by augmenting their workforce.

In many ways, manufacturers are at the mercy of the inventory that comes onto the shop floor. This is the case for manufacturers who process raw materials, as well as for manufacturers who work closely with Original Equipment Manufacturers (OEMs). Detecting variability early is an important part of maximizing value throughout the value stream, and building relationships with trusted suppliers.

Reducing variability from materials means augmenting workers with the tools they need to detect non-conformances before they move downstream. Here, augmentative technologies amplify workers’ ability to determine whether or not materials meet designated quality standards.

Some examples of this include in line quality checks, which use IoT connected devices like scales, calipers, and cameras to augment workers’ ability to identify quality issues before they turn into scrap and rework. Other examples include computer vision, which can work with operators to detect variations in material, as well as missteps at the process level.

Equipment degradation is a fact of manufacturing. In the course of a machine’s lifetime, natural wear-and-tear will introduce variations in output. While equipment degrades on different schedules according to local conditions, advances in sensor technology and predictive analytics have helped improve the ability to pinpoint causes of nonconformance. The key to reducing variability from asset performance is to understand machine usage and part lifecycle from a holistic perspective.

To reduce variability from equipment performance, augment engineers with the machine monitoring data necessary to implement a maintenance schedule sensitive to local conditions. Without the right data, engineers are effectively guessing when and whether a part needs to be replaced. Without a granular perspective of machine performance over time, as well as a documented history of machine usage, there’s no way to isolate a root cause. Understanding variability in equipment performance requires understanding an asset’s human and mechanical failure modes.

Augmentation through advanced analytics enables engineers to reduce downtime, extend the life of assets, and optimize production.

Human performance is perhaps the greatest source of variability in manufacturing. However human variability needs to be understood in context. According to recent research, the vast majority of human errors in industrial contexts aren’t necessarily the fault of individuals. Rather, they’re problems with work systems that put human workers in “error likely” situations.

In other words, smart work stations augment workers by guarding against the most common contributors to variability–overly complex tasks, multiple demands on attention, unintuitive workflows, insufficient training, and poor management.

The best way to reduce variability in human performance is to augment workers with technology that simplifies the tasks in front of them.

For example, digital work instructions can guide workers through complex tasks, simplifying procedures, like high-mix assemblies and intricate machine changeovers, thus reducing the number of variables they need to manage. Rich media like photos and videos, IoT devices, and interactive elements all help to streamline manufacturing work, making it easier for operators to perform each step correctly. This helps them focus fully on the value-add elements of their job.

Ultimately, how a manufacturer chooses to augment their workers will depend on the nature of their operations and the frequency of certain types of errors.

Reducing variability requires two things. 1.) Understanding that human errors are often structural errors, and 2.) understanding how technology can help enable workers. Whether it’s material, machine, or human variability, there are augmentative solutions available to help.