For the next 19 years, the United States will see 10,000 baby boomers retire every single day. That’s nearly 70M people leaving the workforce. Across the industrial sector, there are few resources to effectively capture the institutional knowledge and expertise of these workers before they retire. And, what little is captured has not been effectively disseminated to the broader organization. Therefore, new hires joining the workforce do not receive adequate on-the-job training because their opportunities for mentorship are drastically reduced.
The emergence of new technologies, such as robotics and automation, are also contributing to the skills gap. These emergent disruptions are changing the way jobs look and are requiring workers to have higher levels of digital literacy. According to the World Economic Forum (WEF), 60% of US-based companies state that skills gaps in local labour markets are preventing them from successfully implementing desired technologies.
As such, there is a growing need to reskill and upskill these employees to not only ensure displacement is minimized but to maintain the human resources needed by industrial companies to grow their operations. Combined, these factors are creating a significant demand for greater on-the-job training.
To put this in perspective, by 2030, the global manufacturing industry alone faces a labour shortage of 7.9 million and unrealized output of $607.14 billion.
All that being said, how do we start addressing these looming numbers? I believe the answer lies in machine learning (ML), a subset of artificial intelligence (AI).
Accelerate Adoption of Best Practices
With a workforce that is quickly retiring, it’s imperative we find a means of accelerating the adoption of best practices. This is made possible with ML. We’re able to capture notes and processes, and disseminate these to the broader organization.
Imagine walking up to a machine and immediately knowing its peculiarities without any trial and error.
For example, the ML algorithm would inform every worker not to remove a cap on an industrial HVAC unit’s coolant tank because it leaks. The algorithm would then guide them through the appropriate next steps. Previously, this knowledge was only available to a handful of workers who were maintaining the specific piece of equipment where a note was originally written. However, with an intelligent personal agent, this is available to everyone working on the machine, regardless of location.
In this scenario, the ML algorithm amalgamates the worker’s actions and information they’ve inputted to the software, identifies best practices, and makes it available to all workers. Now, every worker has unprecedented access to continual on-the-job training.
ML for continual learning and training
According to WEF and the Boston Consulting Group, employees and job seekers around the world most value learning, training opportunities, and career development. These are ranked higher than “their job security, financial compensation and the interest they find in their day-to-day job.” With an intelligent personal agent, employees can have a digital mentor in their pockets 24/7.
Using ML techniques, the intelligent personal agent sifts through all the potentially relevant information that may be useful to an employee in a given moment and then presents them with only the most relevant data point tailored to their unique expertise. For an experienced technician, this may be very minimal; for a novice technician, this could be as much as step by step guidance. Being continually provided with on-the-job training will gradually increase an employee’s skill level, effectively upskilling the employee. At the same time, it also enables employees to cross-skill, or develop skills horizontally. An application installed on their device of choice can empower all employees to reduce their time to proficiency and increase the supply of “ready to deploy” experts.
When considering what the future of work will look like with AI, I think it’s important not to view it as a zero-sum game, where either humans or machines win.
AI will become an important component in many, if not all, jobs. But we shouldn’t forget the significance of uniquely human skills, such as creativity, judgment, critical thinking, and curiosity. These are not likely to be automated or programmed any time soon.
As we continue to incorporate AI into the workforce, we should look to maximize the value of both our uniquely human skills and embrace what machines do best – repetitive tasks and drawing insights from copious amounts of data. In this way, we’ll be on the right path to successfully augment the human, making them more productive and efficient.