Eddie Amos is the general manager and vice president of industrial applications at GE Digital. Since joining GE Digital via its acquisition of Meridium, Amos has focused on digital products ranging from manufacturing solutions to Asset Performance Management (APM) applications and other technologies that digitize operations and unlock productivity.
Q: What technology has been influential in transforming industrial plant operations over the past year?
Amos: For manufacturers, mitigating equipment failures and product loss is critical to both the bottom line and the achievement of business goals. And according to LNS Research, asset failures remain one of the top three causes of accidents that result in safety issues or pollution. Each asset, from the smallest valve to the largest locomotive engine, plays a role in powering manufacturing operations. Now, digital transformation has enabled these connected assets to provide valuable data about their performance. This is why asset performance management (APM) technology, which manages the vast volumes of data from connected equipment, has been so influential for processing organizations. APM brings more than analytics to maintenance practices to inform intelligent strategies. While IT software could run analytics on a turbine’s data and generate basic performance metrics, the knowledge that a well-protected, $10 million turbine is 98 percent reliable isn’t that useful when it’s a small valve that fails and shuts the entire plant down. APM has filled a necessary void to provide industrial operators with real-time visibility into the health status of equipment and the enterprise as a whole.
Beyond performance transparency and risk-reduction, APM enables organizations to reduce long-term total cost of ownership, improve safety for workers and the environment, and limit disruptions to operations. For SABIC, one of the world’s largest chemical companies, APM improved the average failure rate for pipes from 172 days to more than 2,125 days, a 1,135 percent improvement. The company also saw a reduction in both leak rate and the total number of failures. In 2017, organizations came to recognize that if an asset is connected, it needs to feed information into an APM system just as it would a historian a decade prior.
Q: What technology is just on the horizon?
Amos: One field that has taken root in entertainment and sports that we can expect to creep into the industrial world is augmented reality. As most manufacturers know, equipment is not always easily accessible. Some machines have hundreds of parts, each embedded inside layers of the strongest materials, and others operate within harsh environments. For decades, equipment location and design have created arduous manual inspection processes, requiring hands-on testing to literally look “under the hood” and identify any problems. With advances in augmented reality for mobile devices combined with the performance insights from APM, technicians can now perform in-depth inspections based on real-time information without having to take the equipment offline and physically analyze it. Augmented reality technology will minimize the risks of infrequent inspections and the risks a processing environment can pose to workers. As a result, companies can maximize efficiency by keeping equipment online with less disruption.
Q: How will new innovations drive change across industrial organizations and at what pace?
Amos: Industrial companies are entering a new period of change marked by unlimited opportunities, but before taking advantage of these opportunities, they should consider where they are sourcing new technologies and strategic recommendations from. Not every “industrial internet of things provider” truly understands the demands of industry. As the technology market becomes crowded, manufacturers must remember that the industrial environment is not a playground for lessons in trial-and-error. For food and beverage, for example, a company’s reputation directly correlates with its product quality. Faulty equipment or machine failure can negatively impact both product quality and brand image, and a recall could mean the end of a business. These organizations shouldn’t be discouraged from embracing technology advances at the same pace as other commercial markets, but they should carefully vet new processes and systems with those who have experience in every stage of the industrial life cycle and understand the complex network of operational technology.
Q: As plants become more connected, how will they effectively manage growing volumes of data?
Amos: While I believe digital transformation is no longer a choice for manufacturers, industrial organizations have historically been slow to adopt new technologies. A recent report found that while these companies recognize the upside of the IIoT, the majority of industrial respondents — 79 percent — say they do not have a mature plan in place to reap the benefits of this connected network. To effectively manage data and leverage key insights, APM is powering digital twins across processing facilities. The digital twin is a dynamic, virtual representation of a machine that is informed by a diverse amount of design, manufacturing, inspection, repair, operational and business data. The digital twin enables companies to draw comparisons between the machine’s current state and its optimal performance to better understand, predict and optimize plant operations.
Eventually asset data can be anonymized and shared industry-wide for a larger data sample as the power of statistics is limited in organizations that run a set number of assets. A broader dataset will help establish more accurate benchmarks and provide better performance insights.
Q: Will technology influence the types of job skills required for industrial organizations?
Amos: Companies are embracing IIoT to drive better business outcomes, but they still aren’t making the most of the data they collect. This is why skills like data science will become more in-demand, but it doesn’t mean every employee should be a data science expert. In fact, we can anticipate an increase in “citizen data scientists” who have some domain knowledge and a surface-level understanding of machine learning algorithms, and can apply those algorithms with different techniques to evaluate data results. Machines will do most of the work for us, and we will need a few brilliant minds to design the algorithms behind them, but most users who interact with these systems will require only limited functional knowledge of data science to reap the rewards of digital transformation.