Automation’s success can be attributed to the value it delivers. As technology has progressed, manufacturing automation has proliferated from basic control on the plant or factory floor to today’s advanced applications, thus opening the door to automation's vulnerability. This makes organizations increasingly dependent on properly functioning applications. Automation has its own dependencies, which must be monitored and maintained.
For example, successful automation depends on accurate, reliable data. Data for automation applications comes from various sources. Organizations must be able to maintain the integrity and flow of these data. As the reach of automation expands, so does the challenge of maintaining the data. Companies must ensure a mature culture around data management to ensure the gains that automation promises are realized.
Automation Increases Profit
While it’s not always easy to quantify in terms of “dollars and cents,” the initial proliferation of automation can be largely attributed to increased profits. Automation enables companies to produce goods at lower cost. It leads to significant economies of scale, which is critically important in capital-intensive industries. Automation enables companies to reduce labor costs and associated issues (including the potential for costly lawsuits and disruptive labor strikes), reduce insurance costs, reduce environmental compliance costs, and increase operational performance and safety.
Automation also enables a greater economy of scope. This means that one plant or factory can produce a greater range of goods, which is becoming increasingly important today for market competitiveness. In the 1950s, 60s, and 70s, the goal was to produce standardized goods as inexpensively as possible. Now, consumers are looking for more customized products. Today’s e-commerce applications allow consumers to customize the size, look, and functions of their clothing, refrigerators, and automobiles, and modern automation enables manufacturers to produce these in an economical manner. The days of picking standard products and models from the assembly line are quickly fading.
Automation can also enable shorter lead times, quicker delivery, and more efficient use of stock and cash flow. It’s no wonder that automation has permeated so many aspects of business.
Initially, automation started with the simplest repetitive tasks on the plant or factory floor. Thanks to computers and networking, automation has expanded into roles that require greater cognition. What was once considered within the realm of science fiction or, at best, academic theory is becoming a reality. Today, automation is permeating up and across all areas of a business; increasingly via the Cloud.
This also means that more and more of an organization’s operation depends on reliable, accurate, and precise automation systems that extend beyond manufacturing to supply chain, accounting, human resources, compliance, and sales, and marketing.
Facing Automation’s Vulnerability
While all automation systems depend on accurate and valid data, manufacturing automation systems are especially vulnerable here. “Garbage in, garbage out,” remains as true today as at any time in the past. Whether due to a worn cam, a bad sensor, poor data compression, bad algorithms or incorrect manual input, effective automation depends largely on the quality and timeliness of the inputted data.
This became particularly obvious when implementing advanced process control (APC) applications. As a wider range of advanced process control applications became more common, these applications would often suffer performance degradation as basic control loops would begin to oscillate. Transmitter calibration would drift, valves would stick, tuning dynamics would change, and the APC would eventually be turned off by the operator. Processes need to be created to maintain the APC and its benefits.
Today, ARC Advisory Group sees many organizations piloting machine learning and other AI-enabled applications for analytics. One of the complaints across the board centers around data quality, which typically requires hundreds of hours of high-dollar labor to address. This wasn’t a big problem in the past because most data resided largely unused in the plant historian. Since many organizations don’t have a uniform method of managing data integrity across the enterprise, rolling out new solutions can be problematic. Just like APC, these new applications are also susceptible to performance degradation over time.
Avoiding Automation’s Vulnerability
As with APC, vigilance is required to maintain the benefits initially realized by analytics applications. This vigilance starts in the field and travels throughout the organization. Because automation is becoming so far reaching, the data sources are varied and disparate. Processes and organizational accountability must be put in place to maintain the gains and the hard work initially invested. Bad habits of the past that lead to problems with the data must be rectified. Interestingly, some of the new technologies will help here by monitoring data sources and alerting users to a problem.
If you would like to buy this report or obtain information about how to become a client, please Contact Us
Keywords: Automation, Maintenance, Safety, Data Management, Analytics, Machine Learning, ARC Advisory Group.