Manufacturing is one of the industries facing a massive labor shortage, especially in terms of maintenance staff, due to high rates of resignation and delays in hiring skilled talent. In this article, Chad Toles, VP of business development, Uptake, how AI can address the issue.Â
While the U.S. labor shortage plagues multiple industries, manufacturing seems to be especially challenged by a lack of personnel. Not only do manufacturers struggle with high rates of resignation, but recruiting young talent and quickly onboarding new hires continues to prove challenging in the industry. In short: manufacturers are seeing employees head out the door, and there aren’t many coming with the right experience to replace them. According to a Deloitte studyOpens a new window from May of this year, as many as 2.1 million unfilled manufacturing jobs may exist by 2030, costing the U.S. economy as much as $1 trillion.
As in so many areas of business, AI has found its way into factory maintenance. For many companies, AI-powered asset performance management (APM) is addressing the staffing issues production managers are wrestling with now. By putting maintenance-focused AI on the plant floor, operations managers will be able to stretch their technical staff and deploy their skills more efficiently. This will do more than extend existing resources; it will also allow a facility’s maintenance workers to be assigned more strategically before a major disruption occurs.
When applied with a CMMS (Computerized Maintenance Management System), AI-powered APM sets proactive priorities and timelines for machine maintenance. AI uses data from machine sensors, repair records, production schedules, and original equipment manufacturer recommendations to forecast when a machine will likely need to be serviced.
Proactive maintenance can reduce catastrophic failures and anomalies, eliminate unplanned downtime, and enable teams to spot issues before they become costly operational problems. In addition, machine learning makes it possible for AI to learn the quirks and variations of a particular process and application over time. It becomes smarter, providing a level of awareness and insights never before uncovered. Less time spent on avoidable repairs means skilled personnel is able to attend to higher-efficiency tasks, like training new talent.
See More: A Call for Governance and Responsibility in AI
Stronger Data Management Leads Keeps Maintenance Teams Efficient
To achieve these labor utilization gains, proactive maintenance systems must be able to take full advantage of operational technology (OT) data. Most companies’ IT infrastructure and data capture practices were established before the AI era. Some of the limitations can be overcome through data normalization and cleansing. Scaling OT data, however, requires additional restructuring.
In decades past, most companies maintained discrete business units that resulted in data silos. Even today, on-premise OT systems like data historians often restrict access to users beyond an individual facility. These walled-off sources of data left blind spots, masking areas of leverage and opportunity. Furthermore, industrial data tends to live in wildly disparate systems. Time-series data may have inconsistent timestamps, for example, and data is frequently missing or filled with errors. As a result, data scientists can spend the majority of their time just cleaning, structuring, and enriching raw data, so it is AI/ML ready.
Correcting this situation requires a management strategy that makes OT data legible to maintenance systems. AI also facilitates OT data preparation. AI-enabled label correction can turn uncategorized or improperly categorized work order data into accurately labeled information based on a training set of high-quality data. In one case, a freight railway company was able to flip its set of work order data from 43% unusable to 93% clean using this technique.
A suitable solution allows teams to better oversee the lifecycle of asset utilization through data-based planning, optimization, execution, and tracking of preventative maintenance activities. It can also significantly improve deployment efficiency and accelerate value realization from years and months to mere weeks.
Connecting sensor data from industrial machinery and processes, maintenance and repair systems, and manufacturing execution systems can help realize the smart manufacturing future so highly sought after. It will allow manufacturers to fix problems before they happen, optimize maintenance schedules for cost and performance, and increase productivity and quality without putting asset life at risk. This allows your available labor force to remain on value-added tasks rather than spending time on avoidable repairs and data management. Not only will it increase your company’s efficiency, but it will add to worker satisfaction and engagement.
See More: 6 Data Management Trends to Watch for in 2022
AI Equips Maintenance Teams With Data-backed Support
AI-powered asset performance management can improve labor efficiency, help teams stay on top of productivity, and reduce overall maintenance costs. It can better balance low-reward maintenance tasks against high-reward tasks without incurring additional risk. Further, this enables departments to reallocate savings to recruitment programs or apprenticeships that build a stronger, more effective workforce while helping to eliminate future shortages.
Fulfilling the promise of AI to help extend staffing, however, will require a foundation of data integrity plus expertise in the proper use of AI/ML. With the right amount of technology, expertise, and corporate will, it will be possible to address challenges with machine failures and technician shortages. Applied correctly, AI will be a key contributor to operational excellence.
Did you find this article helpful? Tell us what you think on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We’d be thrilled to hear from you.