We are sacrificing learning in our quest for productivity. If we do nothing about this, millions of us are going to hit a brick wall, as we try to learn to deal with AI.
So says Matt Beane, Assistant Professor in the Technology Management Program at the University of California, speaking at a Ted Talk earlier this year. Beane references his research in healthcare, where he found that practical skills development was in danger of being undermined by increased automation and an over-reliance on experienced hands.
For most professions, the majority of learning how to actually do the job is on the job. This ability to learn on the job, to work alongside machines and develop new and existing skills, is what Beane has observed as being sacrificed. The human might not be able to do things as fast as AI and machine learning, but it must have the competence to step in where required.
In the short term, the acceleration of AI may not be such a huge problem. It reduces risk of error and is therefore more likely to keep customers happy. In the longer term, this approach will almost certainly lead to huge skills gaps and perhaps a misplaced reliance on automation.
The impact of rising automation
Algorithms and AI are widely tipped to impact a wide swathe of jobs. A recent UK ONS report revealed that around 1.5 million jobs in England are at high risk of some of their duties and tasks being automated in the near future. The Pew Research Center also recently revealed how automation will impact long term job security, with just 14% of adults believing they will have more job security by 2050.
It's a recurring theme across industries. In field service and maintenance, there has already been a change. Growth in sensor deployment has led to an increase in data analytics and the remote management of devices via IoT networks. This has had a knock-on effect in terms of job roles and skills. Businesses have, as a result, become more equipment focused, building customer knowledge as well as machine performance around automated data.
Field service and plant-based maintenance teams have had to adapt, and it has been to their advantage that the emphasis within organizations has shifted. Service is no longer seen as a cost to the business. The ability to provide intelligence on products and customers, and in many cases be the front line for businesses, means service is now strategically important.
What about the workforce?
For many businesses though, this has led to employment issues, especially as the workforce ages. Knowledge loss is an increasingly common problem. According to the Service Council, 70% of service organisations say they would be burdened by the knowledge loss of a retiring workforce in the next five to 10 years, while 50% claim they are currently facing a shortage of resources to adequately meet service demand. Automation is great, but it will only go so far to help.
Businesses need skilled service techs that not only understand the products and machines they maintain and support but can also understand and work with data. Rather than customers asking if a technician can come and fix their equipment, they will be asking what the equipment is trying to say about its performance. This requires service technicians to understand data in context, in the real world, and translate it into customer speak. It demands planning, skills development, and an understanding of the machines to improve service in the future.
Interestingly, the TSIA recently found that half of all field services organisations don’t have a formal career path in place for their field service engineers. This, in my view, is a huge point of unnecessary commercial risk. These organisations are not doing enough to prepare younger service techs for a mixed reality future - one where they will have to work more closely with digital technology and machines than any previous generation. It won’t happen by accident.
The need for human diagnostics strengths
There is certainly a need for an integral ‘system of record’ that captures accurate data about equipment ‘as maintained’. The need for this type of database, showing how equipment looks right now, enables service technicians to understand the context of what equipment data is telling them. While automation can create alerts to problems or potential issues, the service tech will still need to know how to solve those problems quickly and efficiently.
From reading the data correctly, understanding how to correct issues, source parts and manage customer expectations, the fundamentals are not all that new. But as machines evolve with more in-built automation and data-driven analytics, there is a danger that businesses will over-rely on automation, letting their human diagnostic strengths lapse.
As Matt Beane suggests, this could be a problem. Humans need to be able to speak the same language as automated machines, but these machines should not block skills development. In service maintenance terms, AI and automation are not and should never be considered a human replacement. Instead, AI should help humans work better, for example in triaging service calls to assess which can resolved remotely, while talent management should emphasize the new digital skills that the field workforce needs. If anything, AI and automation are opening up the industry and creating a brand-new field of opportunity.