5 steps to automation success in metal manufacturing

FIGURE 1. A cobot loads parts onto a flat-part deburring machine, and a cobot tends an electric press brake. Challenges vary with the application, but if an operation is somewhat simple and tedious—that is, one that employees wouldn’t mind going away—it might be a good candidate for automation. Images: Mid Atlantic Automation

Reshoring efforts, supply chain challenges, increasing wages, an aging population of skilled workers, a generation-low unemployment rate—metal fabrication’s challenges seem to multiply relentlessly by the day. And they continue to grow and amplify one another, no longer just threatening a fabricator’s lead times and margins, but its fundamental ability to maintain business continuity and customer relationships. Right Angle Brackets

5 steps to automation success in metal manufacturing

Proven, widely adopted automation solutions can alleviate some of these stresses, yet many organizations are reluctant to embrace even relatively straightforward applications. Why won’t they take the first steps? It might stem from a natural resistance to change and fear of the unknown. Managed properly, change forces organizations to learn and grow, and it’s indispensable for an organization to remain relevant and competitive.

Failure to change can be disastrous. Consider Kodak’s story, a well-known cautionary tale. Its roots going back to 1901, the one-time photography and film giant had been a market leader for decades. But as digital photography proliferated in the early 2000s, Kodak chose to remain focused on its core business of film. Digital cameras rapidly demolished this market, and by the time Kodak reacted, it was too late. The company declared bankruptcy in 2012.

Automation is rapidly becoming a necessity for businesses to be able to keep up with customer demands. Will your organization embrace the change of automation, or will it become the next Kodak?

Automation might appear intimidating, but it shouldn’t be, as long as you have the right focus. To start, try focusing on five common-sense steps. Followed well and repeated in a virtuous cycle, they help build a broad framework for metal fabrication automation—one that goes far beyond a single process or workstation.

As with any journey, the first step on the path to automation is determining where you’re headed. Set expectations and make them clear. What does your organization want to accomplish with automation, and what are the impacts? For instance, is the intent to automate incrementally or on a large scale, and how will automation impact workflow for engineering, planning, and production? Who will be the designated automation champion?

How do employees perceive automation? Many see it as a job killer. In truth, though, workers who become adept in robotics and automation increase their skill set, enhance career opportunities, and become more valuable to their organization. Companies must identify and support automation experts in engineering, manufacturing, and on the plant floor.

Finally, what are the financial implications? Financial success lies not only in cost savings, but in expanding a firm’s capabilities to reduce lead times, ship more products out the door, and bring in more revenue.

In a nutshell, you must plan for automation. It won’t solve every problem all at once, but it can solve the right problems very well, and it can create (often unanticipated) opportunities for innovation, including additional career opportunities. Automation can save money, but even more importantly, it brings more revenue. To make this happen requires focused and empowered program champions who work to achieve clearly defined goals; communicate clearly; and conduct consistent, thorough reviews of the progress being made.

Rank targeted processes based on risk and their overall impact to the business. Automation solves the right problems very well. The trick is to select the right processes to automate in the right order and match it to the optimum automation technology.

FIGURE 2. This matrix shows the difficulties of automating specific operations at a metal fabricator. The matrix will vary depending on a company’s mix of products, processes, and available resources.

For instance, industrial robots are well adapted to large runs of uniform parts and have significant initial costs. Simple operations with little variation are well suited to relatively inexpensive devices. In between, with small to large volumes containing significant part variability, flexible systems such as adaptive collaborative robots (cobots) balance cost and complexity.

In the past, the decision to automate often depended on a simple return on investment (ROI) calculation. “We make X number of this part per month at Y cost and can save this much per part. Capital cost is Z, and the payback period is this many months.” In today’s environment, however, the picture is more complex. ROI calculations are helpful, but often the effect of a bottleneck caused by an idle machine is much more impactful to a company’s bottom line. If a large end item cannot be shipped because of relatively inexpensive missing fabricated components, then the real payback must be viewed in terms of the revenue loss caused by the bottleneck—a financial effect that can be many times what a simple ROI calculation would show.

An effective approach involves ranking impactful processes based on the overall needs of your business. Every factory is different and will weigh variables differently. Among the most common factors are overall cost as a percentage of revenue, impact on customer lead times, difficulty acquiring skilled operators for the machinery, workflow bottlenecks, unacceptable quality levels, safety issues, workstation idle time, complexity of integration, and overall process pain points. If many people involved with a work element really wish it would just go away, then it’s probably a good candidate (see Figure 1). Charting the variables in a simple graphic that ranks overall business impact and difficulty of implementation can crystalize the decision-making (see Figure 2).

Ranking processes like this helps narrow the selection to a small number of potential targets. At this point, decision-making is as much art as it is science, so it’s generally not useful to overanalyze. Pick the process that appears to have the best balance of implementation complexity and business benefit, and start there.

When doing so, though, be sure to evaluate your existing engineering and automation capabilities honestly. If you have some internal resources and want to self-integrate, pay special attention to the integration difficulty. Don’t try to do everything at once, but do try to keep in mind the level of automation you are willing to start with. It might be as simple as a single-axis pick-and-place device that moves workpieces from point A to point B. At the other end of the spectrum are highly customized industrial robots that are appropriate for high-volume output of stable workpieces. In between are devices such as cobots that are quite flexible and relatively simple to adapt and program with the proper tooling and application support.

For any type of automation, never overlook proper safeguarding. Automated systems must be deployed safely, with safety costs incorporated in the budgeting stage of project planning. Of course, the complexity of the chosen process, the time and capital expenditure, and the volume profile of the chosen parts being built all affect the overall decision.

Automation usually doesn’t replicate a manual operation. Instead, planners re-engineer the process for automation, always in the context of overall manufacturing flow.

A good first step is to generate a simple map of manual processes (see Figure 3). Nothing exhaustive, the map just needs to describe the elements that must be addressed. After you draw your map, analyze the factors that affect it. Are the inputs (incoming parts, for example) stable in design and repeatable in form? Is the existing process stable? Is the existing equipment capable? Are present quality levels acceptable? Is the ongoing human interaction with the process predictable? Does the existing process represent any safety or health risks? Resolving these fundamental issues first helps build a solid foundation.

Once you address the fundamentals, generate and publicize a clear and concise list of the critical goals and corresponding deliverables for the project. Set up a periodic review process to make certain goals are still relevant and on target. Never lose track of the simple fact that automation doesn’t take the place of existing tasks but instead requires everyone to rethink the entire manufacturing process. Bolting a good robot onto a poor process never works and can be very expensive. The upside is that building effective automation often exposes and forces resolution of underlying (and often unrealized) problems.

Another critical factor is often the most difficult to address: an honest and realistic evaluation of available internal resources and skills. Foremost is the designation and empowerment of one or more automation champions. They don’t necessarily need to be highly experienced and skilled in automation, but they should thoroughly understand your existing processes and your automation project goals. In addition, they must be allowed to focus on the project rather than be forced to treat it as a once-a-week diversion.

Automation might appear intimidating, but it shouldn’t be, as long as you have the right focus.

Champions can cede some level of project control to outside integrators. Still, they must balance that control against an integrator’s scope of experience, technical expertise, and the potential of a shorter ROI. Ask a prospective integrator for references from existing customers and examples of their completed work.

Remember, engaging an integrator does not replace or alleviate the need for an internal automation champion. Project implementation, feedback on progress, and focus on completion schedules require clear communication channels with any integration team, and a designated individual needs to drive this process.

Once you’ve outlined the project deliverables, empowered an automation champion, and defined the integrator’s role, it’s time to review the project deliverables with laser precision. Demand that they be clearly defined and documented, and formalize a periodic review. Clearly designate the roles of the automation champion, the implementation team, and the integrator. Document these roles and review them in a structured way.

Run a final check on what you expect from the project, that the deliverables align with these expectations, and that they are realistic and achievable within the allotted time frame and budget. Don’t overreach.

Finally, vigilantly avoid the traps of unrealistic expectations and project creep. Your first automation project will not solve every problem in your factory.

Continuously monitor the process’s performance and impact on the rest of the factory. One of the most fulfilling results of a successful automation project (beyond enhanced revenues and ROI) is the attendant growth in institutional knowledge. Even a simple first program provides extensive learning in how to choose a project and set goals, how to review candidates for future automation, how to pick the right equipment and individuals to get the job done, and how to predict timelines and financial benefits.

Finally, document ongoing automation successes as well as areas that need improvement. These data-based insights help build the foundation for future automation projects. You’ll also see where specific automation strategies could be replicated elsewhere.

Always document with candor and clarity, and celebrate successes. Pick the next project as carefully as the first using what you now know to take the next, more productive step in your automation journey.

No level of technological wizardry can solve every headache in your factory, but taking a disciplined approach to automation can help you win. Any initiative will only be successful with a clear plan, the right people, thoughtful preparation, and intelligently developed expectations.

Automation is an opportunity to develop a clearer understanding of your operations, drive innovation, develop valuable processes and institutional knowledge, identify and strengthen capable people, and build your bottom line. Most of all, managed automation can better serve your customers and allow your organization to remain relevant in today’s rapidly evolving business climate.

FIGURE 3. A simple process map, which need not be exhaustive, details the basic steps of a specific manufacturing process that the automation must address.

Mid Atlantic Automation, a division of Mid Atlantic Machinery

See More by Joshua Mayse