The Impact of AI and Machine Learning on Field Services

James Hoshor | October 22, 2018 | In the News Mobile Strategy

As we’ve discussed in previous blogs, it’s just a matter of time before every company uses AI and machine learning (ML) to achieve key business drivers. This post describes how the Field Services industry is using AI/ML and details the impact of AI and Machine Learning on Field Services.

In a recent Service Council Field Service Benchmark Study, the top three metrics that service organizations are working to improve are:

  • Workforce Productivity: schedule optimization, street-level routing, truth-based appointments, mobility, capacity planning
  • Customer Satisfaction: narrow appointment windows, real-time status update messaging
  • First-Time Fix: augmented reality, knowledge management tools, analytics

These are pretty common metrics. But introducing AI/ML changes the game, enabling predictive field services for maximum positive impact.

Let’s examine the Field Services Lifecycle (FSL) to better understand where and how AI/ML delivers value. We’ll separate the FSL into 3 areas: Prior to-Field Service Engagements, During-Field Service Engagements, and Post-Field Service Engagements.

Impact of AI and Machine Learning on Field Services

Prior to Field Service Engagements

Let’s start by looking at service planning to meet customer service demand. Several variables should be taken into consideration when planning services based on demand. AI and machine learning can make a big impact here by using a combination of heuristics and predictive techniques to establish which underlying variables have historically been the most important factors in predicting actual demand. This enables Field Services organizations to effectively plan resources, mitigating overtime cost and minimizing the need for having to recruit contractors last minute to meet customer service needs. Applying ML to historical data captured from service planning and customer demand forecasting can help realign territories to ensure adequate coverage is provided to meet customer demand.

Preventive maintenance is another area where AI/ML can make an impact. Many factors come into play here, such as: how often equipment is used, age of equipment, expected lifespan, facility conditions, etc. Despite the best intentions, the practice of preventive maintenance is inefficient. Preventive maintenance is setup to be performed on a regular basis—whether needed or not. Organizations don’t take into consideration actual usage of equipment and sometimes don’t identify equipment defects. This can result in replacing parts that don’t need to be replaced or sending out a Field Service technician to service equipment that’s rarely used and therefore doesn’t need servicing.

We recently worked with an organization that employed over 600 Field Service personnel, each with regular, defined service routes and equipment to check and maintain daily. AI/ML enabled them to identify and prioritize which equipment should be regularly serviced and which could be serviced less frequently. It’s an inconvenience when a Field Service technician has to perform an unnecessary fix or waste time driving to a site that doesn’t require work. But when this happens regularly across tasks, jobs, and technicians, costs can seriously add up.

Our client got back 2–3 hours per day by letting Field Service personnel focus on critical tasks and priorities, and eliminated the high cost of driving to locations that didn’t require servicing, including a significant reduction in overtime.

Using AI/ML lets preventive maintenance becomes predictive—by taking actual equipment condition into account when determining repair/replace schedules. Organizations that have implemented AI/ML use sensors to monitor equipment performance, capture the data and analyze it to identify any abnormalities. When performance falls below specific thresholds, work orders are automatically generated and sent to Field Service technicians to perform.

80% of technical experts across the industry believe AI enhances workforce skills and increases work efficiency.”

During Field Service Engagements

One of the ways ML positively impacts Field Services is by providing accurate estimations of task time and resource utilization planning. By identifying the steps required to perform a task, ML accurately determines the actual time required to perform the task (versus providing a service window). This helps with future scheduling of service requests, providing better response time for customers and improvements in service planning. As more data is captured over time, organizations may continually improve their scheduling systems, further refining the actual time required to perform service tasks. AI can also be used to assess the factors that impact a high first-time fix rate by using historical data to determine which parts, tools and resources are required to ensure a successful service engagement.

The combination of data and AI/ML optimizes field services delivery and operational efficiencies in the following ways:

Dispatch & Routes: Field Service technicians spend a lot of time behind the wheel. When routes aren’t properly optimized, windshield time wastes service organizations significant money each year in lost revenue, fuel cost, and technician pay. AI/ML can help solve this:

  • Dispatch—AI algorithms can significantly improve worker efficiency by automating assessments of route, customer history, and technician skill level to instantly identify the ideal field service technician for each job. Technicians could be informed of dispatch requests via push notification.
  • Routes—Many situations impact routes, including real-time traffic changes, construction, obstruction, etc. Not all service-route optimization-scheduling software accounts for these variables, but AI can assess all available data in seconds to ensure the field service technician is informed of the most optimal route.

Predicting First-Time Fix: AI/ML identifies service queue patterns, notes how similar patterns have been handled previously, and identifies better options, given previous outcomes.

Territory Planning: AI/ML creates optimal service request groupings such that field technician territories are dynamically created on the day of service, rather than using static, and inherently inefficient geographic boundaries.

Post-Field Service Engagements

Every client we engage wants to improve the optimization of its field services delivery. To do this effectively, organizations need to understand what to optimize for. Technician overtime is one area organizations consider first. But without having all the field services engagement data—from planning to execution—organizations don’t fully understand the impact of such a move. Simply eliminating overtime doesn’t always provide the desired results. Technicians may stop working beyond their regular hours, reducing the number of jobs that can be undertaken, especially since jobs that would finish a few minutes past regular hours would be moved to the next business day.

Naturally, this has implications for other service metrics: utilization and customer satisfaction rates will fall because time to service increased (though field service employee satisfaction may improve).

This being said, AI/ML are not a silver bullet or end-all solution. However, AI/ML can help translate these business imperatives into an “optimal” schedule, resulting in better decision making when it comes to service planning and delivery.

The ability for organizations to respond quickly to continually changing business environments and conditions, and make informed decisions is critical. Leveraging AI/ML can help by analyzing data captured throughout the Field Services Lifecyle and enabling the adjustment of service and resource models, resulting in operational efficiencies and—more importantly—higher customer satisfaction.

If you haven’t given AI much thought yet, it’s time to start. Industry-leading companies are already having great success with AI and machine learning solutions. You can, too. Contact us to help your organization develop an AI/ML strategy and roadmap that leverages these emerging technologies to ensure alignment with business drivers for guaranteed success. Our (Anexinet’s) Machine Learning Strategy Kickstart may be just the boost your business has been looking for.

James Hoshor

James is a Senior Mobile Strategist & Solutions Architect for Propelics. He has over 20 years experience in executive leadership, strategic planning, marketing and business development in information technology. For the past 10 years James has worked with many clients across multiple industries, including financial services, insurance, retail and manufacturing, approach mobile strategically to identify and deliver solutions that result in market differentiating solutions and great user experiences.

More Posts

Follow Me:
Twitter