By Paul Whitelam
January 2, 2018
How many times have your dispatchers spent hours trying to create the perfect schedule only to have it ruined the day of service when an emergency job arises or a customer cancels? Suddenly they’re scrambling to reshuffle the schedule they already spent so much time on. And they’re forced to rely on their gut instincts instead of calculated decisions because they just don’t have enough time to consider every scenario.
If only they could have foreseen these disruptions ahead of time, they could have prepared for them. Not to mention, if they had the processing power of a computer, they could make lightning-quick and accurate decisions in case something unexpected does pop up.
Humans may not have the power of clairvoyance or extensive processing, but fortunately there’s technology that does. Machine learning gives computers the ability to learn and make decisions without being explicitly programmed—by recognizing patterns based on information gathered and stored. When used in field service, machine learning empowers organizations to provide better service with predictive insights and data-driven decision making.
Predictive Field Service: Powered by Machine Learning
Think about how frantic your customers are when something breaks and you aren’t able to fix it right away because you don’t have the available technicians on hand. And how many times has a customer complained because their tech didn’t have the right parts to fix their problem on the first visit? In the midst of an already packed day, your team is putting out fires and salvaging relationships with customers.
But with predictive field service (PFS), reactive service delivery becomes a thing of the past. PFS uses machine learning to draw conclusions from data, and allows service businesses to get ahead of disruptions and issues before they happen. It also allows intelligent field service management solutions to anticipate changes and automatically adjust business processes accordingly.
But can you trust machine learning with your business?
You might feel hesitant about trusting a machine to make business decisions for you. After all, how could a computer possibly understand your business, technicians, and customers? But in reality, every decision it makes is based on business decisions your team already made in the past—so you can trust that they make sense.
Though no computer could replace your dispatchers’ experience, instincts, and relations with customers, it can help them make better, faster decisions and accurate, data-driven predictions. Let’s discuss the main advantages of machine learning over the human brain.
1. Machine learning is fast
Machine learning technology can process vast amounts of data in seconds to make predictions. And it can quickly simulate millions of scenarios to determine which makes the most business sense. This kind of processing is beyond the limits of the human brain.
And why does speed matter in field service? Because timing is everything, and even a few minutes lost or wasted at a time can add up to major costs. For instance, a tech might only be ten minutes late for a service appointment, but that puts him behind for the rest of the day. Multiply that ten minutes by the seven or so jobs the tech has that day and it adds up to more than an hour of lost time (and at least one missed appointment).
2. Machine learning is more accurate
Because machine learning uses data and advanced algorithms to make predictions, it’s much more accurate than a human could ever be. While an experienced dispatcher can probably guess that during morning rush hour it could take a technician twice as long to get to a service appointment, a computer could tell them exactly how long it will take and the route that will get them there the fastest.
Machine learning likewise eliminates the “gut feelings” and biases humans are susceptible to. For instance, a long-time dispatcher might believe like they know better than a computer or anyone else about technician skills. And when an emergency job pops up, they send a tech that usually works the fastest, not remembering that this individual wasn’t trained to do that kind of job. A new dispatcher could easily make a similar mistake. They might feel extra pressure to make a quick decision in the midst of a crisis and send the first available tech to the job—without considering their skill set.
However, rather than relying on one person’s memory and insight, machine learning analyzes and learns from all available data. It has every piece of data needed to make optimal business and scheduling decisions. Humans simply can’t store that much information.
Of course, none of this means humans don’t have a role to play. When it comes to making more nuanced decisions or interacting with customers, humans are much better suited. Machine learning also relies on the data that humans provide and input into the system. Think of this technology as an additional team member that can automate tasks and free up the humans to be more productive.
By using machine learning to power predictions and automate the optimal decisions for your business, your team can help more customers per day and handle anything that’s thrown their way. It can mean the difference between a service disaster and perfect customer satisfaction scores.
To learn more about intelligent and predictive technology like machine learning, subscribe to the Field Service Matters blog.
By Paul Whitelam
Paul Whitelam is Group VP of Product Marketing at ClickSoftware, where he works with field service management leaders across a variety of industries. Paul has more than twenty years’ experience in enterprise software, working on both the technical and business aspects of many of the areas that are fundamental to field service such as mobility and sensor technology (Nokia), data management (Endeca), and machine learning, SaaS and GIS (HERE).
How many times have your dispatchers spent hours trying to create the perfect schedule only to have it ruined the day of service when an emergency job arises or a customer cancels? Suddenly they’re scrambling to reshuffle the schedule they alread...