Thirty years ago, an operator might have spent most of a shift walking the floor, feeling the pump casings for vibration or eyeballing analog gauges. Today, however, the same plant is likely half the size and thirty times as data-rich. Its variable-frequency drives modulate load, while fault-detection algorithms flag abnormalities in chillers and a control dashboard tracks thousands of sensors in real time. This progress has trimmed energy use and freed space for revenue-producing functions, yet it also has widened the skills gap. Younger hires arrive fluent in software but light on mechanics; their mentors can rebuild a steam trap blindfolded but may hesitate to trust an optimization routine. The result: mechanics who can fix anything mechanical but distrust automation, all while digital natives who read the data wait to reach for the wrench last.
The workforce math is stark. Nearly 50% of utility-sector workers are projected to retire within the next decade, posing a risk of losing critical institutional knowledge for mission-critical systems and operations. At the same time, 1 in 4 utility executives say skill gaps are holding back adoption of AI and other advanced digital tools. Recruiting replacements is difficult because utilities, hospitals and data centers compete for the same talent.
To help prevent knowledge from walking out the door, many savvy facility managers have begun scheduling structured capture sessions: interviews, screen-recorded demos or even 360-degree videos to document how veterans diagnose issues an automated system might miss. The “knack” for knowing that smell, sound or sequence is unique to each plant and cannot be taught in a book. Consulting such on-demand files, in a searchable library, is a sustainable stand-in for asking a veteran operator a question about a system. A quick clip of an experienced mechanic explaining acceptable chiller sounds often teaches more than a thick manual.
Next, institutionalize blended upskilling by pairing mechanical work with digital validation. For example, after replacing a pump seal and restoring flow, a technician can use the analytics platform to verify that motor amperage has returned to baseline. Combining the wrench with the laptop reinforces the connection and helps operators see software as part of the workflow, not separate from it.
Modern tools, like digital twins, push learning and real-time diagnosis even further. A twin is a live 3-D model fed by real plant data. Operators can test an emergency shutdown on the screen or even be guided virtually by manufacturers. This training is integral for younger staff to rehearse scenarios or for more experienced technicians to preview new technology in a simulated environment.
Trust in new systems, ultimately, remains a hurdle. Experienced operators still run equipment manually to “hear the motor,” regardless of software recommendations. Modern systems help alleviate the gap with building analytics. When an operator chooses to make a manual decision, modern systems will make an informed “ping” on whether it deems the choice accurate or not. Some systems might even display dollar amounts behind the adjustment, to illustrate the potential deficit the operator might be making on behalf of the airport. Small transparent wins with developing technologies build confidence faster than directives.