Customers Driving Industrial Automation Innovation

Image Source: Zapp2Photo/Shutterstock.com

By Darshan Pandya for Mouser Electronics

Published August 11, 2021

Customer needs and expectations are the driving force behind product development and manufacturing. Customers want product options and personalization. Customers want customized products manufactured and delivered quickly, and increasingly, customers expect manufacturers to align with their values. Although the traditional manufacturing goals of efficiency, accuracy, and safety remain relevant, manufacturing environments are trending toward new characteristics, such as being agile, accessible, data-driven, collaborative, resilient, and sustainable. The following examines these characteristics and how they help meet customers’ rapidly evolving needs and expectations.

Agile

Agile systems are rooted in batch production approaches, where the base product is mass-produced, but customization is carried out in batches—say, producing one batch with red paint, another with blue, and another with pink. Agility relates to market changes and what customers want; it changes the concept of producing one or two variations to instead offering those variations as options, taking manufacturing from customizing products to personalizing them. Whereas customization leads to customer satisfaction, personalization has an aspect of customer delight attached to it.

The most advanced systems produce each product as a separate order along the production line, using customer data to drive personalization. Imagine ordering a product during the week of your birthday, for example. The manufacturer might add a birthday note in the box or print “Happy Birthday!” on the packaging to personalize the product. Or, say you have been researching meal delivery services online: A manufacturer might put a coupon in the box for a popular food delivery service. With batch production processes combined with customer data, manufacturers can do a lot to personalize products and not just satisfy customers but delight them.

Accessible

In industrial automation, accessibility refers to two different aspects. First, accessibility refers to connecting a distributed workforce through integrated, secure systems. Such systems have been evolving for decades, but only recently have the technologies necessary converged to enable seamless and secure collaboration. In earlier decades, we saw various components and pieces of technology make collaboration possible, but only now do we have what we need to fully integrate distributed workforces, resources, and services across all aspects of business.

Second, accessibility refers to integrating human factors into system design in terms of ease of use for installers, operators, technicians, floor workers, and others. Designing with humans in mind used to be an afterthought or a nicety. Now, companies realize that the installers, operators, and all the rest are the consumers in these cases. Therefore, considering human factors and applying related guidelines are important parts of design success.

In both cases, cloud infrastructure, with system integrators such as infrastructure as a service, platform as a service, software as a service, and similar concepts are the most significant enablers of accessibility. Teams working in real-time with minimum lag time provide considerable extensibility for the types of collaboration possible. Augmented reality (AR) is a good example. Rather than engineers, technicians, manufacturers, and other stakeholders flying in to install, operate, troubleshoot, or repair a mechanical system, AR enables stakeholders to access data analytics, see the system in real-time, and use visual overlays to make repairs or alterations. The efficiencies related to cost, time, and product life cycle are already significant, and we can expect them to increase the number of AR technologies improved over time.

Data Driven

In today’s industrial automation, data drives many insights. For instance, data can evaluate how a manufacturer is doing in terms of what is most important to the company. Here, value matrices are used to identify the five or six aspects most important in measuring the company’s success, such as speed, accuracy, conformance to standards, conformance to regulations, and customer satisfaction. Using real-time data and data trends over time, stakeholders can see how the company performs overall, performs in terms of its values, and performs compared with industry averages.

Data insights have moved humans from simply responding to manufacturing issues to instead proactively preventing and addressing them. Fluctuations in data can be used to identify when something is about to go wrong and then trigger email messages to responsible technicians, support tickets filed with the help desk, or text messages to supervisors. In some cases, fluctuations can be benign; nonetheless, real-time data and trends can help put humans in a proactive rather than reactive role.

Finally, data can be used to evaluate the industrial automation characteristics discussed here. For instance, data can enable greater manufacturing agility by providing insights about when a manufacturer should start offering a sofa in additional colors or different fabrics rather than manufacturing those instances as product variations. Data can also drive process resilience by minimizing downtime for product variations and ensuring that humans are in the right places at the right times in the processes.

Collaborative (with Robots)

The nature of collaboration is also evolving with humans collaborating with other humans and robots. The goal of using machines and robotics in industrial environments has historically been to shift heavy, repetitive, and dangerous tasks away from humans to mechanical and automated systems. Collaborative robotics (cobots) expands these use cases, however, to include:

• Skills that require years of training for humans to become proficient. Welding is one such skill where more than 100 hours of classroom time and three to four years as an apprentice are required. Even then, human welders typically cannot match the quality and consistency of welds that robots can make. By way of example, even highly skilled welders can weld only a 60cm seam in one continuous motion, and starts and stops can affect the overall quality of the weld. In contrast, robots can weld roughly a 121cm seam in one continuous motion, producing a higher-quality weld. A collaborative approach for welding might have the robot completing steps that require skills and might affect quality. The human would do the tasks that are more intuitive to humans, such as setting up the process and handling exceptions. The idea here is to use the best of both worlds to achieve maximum efficiency.
• Tasks where robots can do part of the work. Imagine receiving components that are fragile or sensitive to handling or that require a clean room. In these instances, using a cobot to open boxes of various sizes and shapes would pose several design challenges and be less efficient or effective than having humans perform such tasks because the tasks are intuitive rather than calculated. In contrast, a cobot could easily meet the sensitive component handling requirements after the box has been opened and perhaps do so faster and more consistently than humans.

Resilient

Resilience is about enabling systems to adapt quickly when variations in product manufacturing are needed. For example, imagine a system that is designed to produce product A. At the start of the production run, someone sets up the system to manufacture that product. However, what happens when variations of the product are needed—a different color, size, fabric pattern, module, or packaging? In less resilient systems, someone would need to manually change the setup to accommodate these variations, which means production downtime and higher labor cost. Resilient systems can withstand such unforeseen adversities and recover quickly. The aim of resilient systems is to account for possible variations that the system might encounter to adapt with as little downtime and human intervention as possible.

From a design engineering standpoint, it’s easy to get trapped in the idea of developing super-complex systems that can do many things and account for even rare scenarios. The more complex the system, however, the more opportunities for failures. A simpler system that handles 80 percent of production scenarios might be better than a more complex system that handles 90 percent of scenarios. Designing for resilience is about finding a balance between simplicity that requires more human time in the loop and complexity that has more failure points and potentially affects other production areas.