When some people hear the terms “artificial intelligence” or “automation,” they may feel a little anxious. They wonder: Are machines going to take our jobs? Can we trust the decisions made by artificial intelligence? To put this reaction in context, according to McKinsey, 51% of work activities can be automated but only 5% of occupations. In other words, automation helps humans do much more with less time and effort, so they can put their skills to better use.
The ideal for field service is that repetitive and time-consuming tasks are delegated to automation, while people (i.e. technicians and dispatchers) are free to do work that requires the human touch. Artificial intelligence can automate and optimize routine scheduling decisions by processing massive amounts of data and making predictions at lightning speed. The ultimate vision for field service automation is called “Zero Touch,” or 100% automation without human intervention. Before we dive deeper, let’s start with some definitions:
Definitions: Zero Touch Automation, Machine Learning, & Predictive Analytics
Zero touch automation is a vision for field service management, where the back office is fully automated and optimized in line with the company’s business goals. This technology leverages a mix of machine learning and predictive analytics to maximize operational efficiency.
It’s important to note that zero touch is a vision that may not be attainable, or even desirable in all situations. In some cases there will always be a need for human interaction, especially in emergency or unplanned situations. But service providers should strive to get as close as possible to zero touch and leave humans to interact with customers, fix equipment, and make non-routine decisions.
Machine learning is a type of AI that gives computers the ability to learn from historic data and improve the quality of decision making without being explicitly programmed.
Predictive Analytics encompasses a variety of statistical techniques from predictive modeling, machine learning, and data mining that analyze current and historical facts to make predictions about future or otherwise unknown events.
What can be automated?
In field service, routine scheduling and decision making can be left to automation. AI can process large amounts of data in seconds to make decisions that would take humans much longer to process. To illustrate, let’s consider in-day scheduling changes.
Though you might have a service schedule a ready a day or week in advance, it’s likely to change on the day of service due to a variety of factors:
- Inaccurate estimated travel time and work duration
- Last minute cancellations
- Wrong or missing parts
- Technician skills not updated correctly
- Lack of geographical coverage
- Weather considerations
- Higher priority work
On the day of service, schedule adjustments must be made quickly, and humans don’t have the capacity to determine to best solution, given all the trade-offs and possible options in a matter of seconds or minutes. That’s where automation comes in handy.
For instance, this technology can accurately estimate travel time and optimize a technician’s route based on traffic conditions. Based on things like customer demographics, customer history, and task type, it can predict the probability of appointment cancellation and automate reminders for higher risk appointments. And with predictive weather feeds it can readjust the schedule to prioritize high priority jobs and reschedule lower ones based on the likely arrival of a storm.
How can you prepare for zero touch automation?
Before your organization is ready for such a high level of automation, there are a number of things to consider. Here are just three examples:
Data Cleanliness and Availability
Without the right data, automation is impossible. If you haven’t been keeping track of the right data, such as technician efficiency, job duration, customer cancellation rates, or customer demographics, you won’t be able to leverage most of the optimization and predictive capabilities of automation. Or you might have data, but it’s disorganized and spread across systems. Sooner than later, determine the tasks you want to automate and begin collecting and organizing the necessary data so you can measure what success looks like.
Machine learning thrives on leveraging large amounts of data to make the most accurate decisions. If your data is spread out on multiple systems, this may compromise the effectiveness of the approach. Let’s say you want to automatically schedule a tech with a particular skill set to all jobs that require that skill set. But if that technician isn’t listed in the scheduling system, they won’t be considered in the automation.
Get the whole organization on board
It’s harder for some people to trust automation, especially with tasks they’ve been doing for a long time. For instance, a dispatcher might reject automated decisions because they think they could make better ones. Likewise, the high percentage of retiring field service techs might be more resistant this kind of change. It’s an idea that will take time to get used to, and not everyone will be on board at first. And finally, decision-makers might feel too overwhelmed to make the huge investment in a system upgrade. Begin by educating everyone on the benefits of automation and emphasize that it’s meant to help them, not replace them.
With all that automation can do, it’s understandable that people may be fearful of it. But it’s important to stress that automation is also meant to make employees’ lives easier. Free of time-consuming routine tasks, technicians and dispatchers are free to handle what can’t be automated, such as adding the “human touch” to customer service.
For more on automation and field service trends, subscribe to the ClickSoftware blog.