Source – forbes.com
Over the past decade, technology has redefined how service businesses approach jobs. From heavy equipment services to residential services, anywhere a clipboard was the traditional method of documentation, mobile technology has been claiming a stake.
Mobile automation solutions have become the norm, giving service organizations the ability to communicate service records digitally. Sharing work orders between the back office and the service technician has become a more efficient, digitized process with near instant transfer of information.
Artificial intelligence is going to be the next revolutionary stride toward improving technician utilization, first-time fix rates and the customer experience. In order to accomplish this, we will need a system that will be able to learn and run a service department effectively.
Over the past five years, machine learning image recognition accuracy has improved, moving from a 30% error rate to a 3-4% error rate. To give some perspective, the average person has a 5% error rate in image recognition.
Thanks to this improvement in accuracy, the technology can quickly recognize features within images for quicker access. This can be a useful tool for searching through asset history and customer records that have been recorded in a digital work order.
However, this current technology has the potential to evolve into an even more powerful service tool. By developing it into an AI system that can recognize diagnostic issues from a live camera feed, it could be used to assist service technicians to troubleshoot issues faster.
The system could evaluate the current condition of the asset, comparing it to the historical records of the asset to make inferences about what the issue may be. In addition, by using this troubleshooting method, the technician is building a more comprehensive asset history that will be available when future maintenance is required.
Work Assignment
Traditionally, in a service organization, a customer calls to request service and an employee will record notes and generate a work order. Next, that work order will be added to a queue of jobs to be completed and assigned to the next available technician.
With a system redefined by artificial intelligence, the workflow will be customized by the capabilities of your workforce. By evaluating data from your organization’s mobile automation solution, AI will be able to learn about individual members of your service team and adapt work assignments to match individual capabilities, maximizing the efficiency of your workforce.
By tracking the amount of time a technician spends on a job site, materials consumed and first-time fix rates, the system will be able to score performance that could be used as a factor when assigning work. Scoring based on these factors, the system has the potential to spot and evaluate trends that may have previously gone unnoticed. Taking too long to complete a job or using excess materials may indicate a lack of skill. Alternatively, completing jobs at a breakneck speed could be a sign of cutting corners.
Scheduling
When AI is determining who should be completing jobs, it would only seem logical that the AI would then decide when that work will be completed. Unlike a traditional dispatcher, an AI-based system will be able to use a multitude of factors to generate an efficient schedule.
To generate an optimal schedule, AI systems will likely be able to evaluate weather forecasts and traffic trends that have been the cause of routing inefficiencies in the past. Based on these factors, the system will be able to allot additional travel time. For example, in Pittsburgh, an AI traffic-light system named Surtrac has been deployed and has reduced travel time by 25%.
As an extra parameter, the system can factor in the preferred visitation times that may be requested by a customer. This helps to mitigate the classic customer pain point experienced when they have to wait around all day for the repairman who was supposed to show up at some point in an eight-hour window.
For organizations that provide services that tend to be on demand and require immediate attention, AI would determine schedules by job location, forecasted completion times, parts availability in service vehicles and severity of the issue. Regardless of the situation, an AI system will objectively assign work in order to get an optimal return on investment.
Maintenance Forecasting
Preventive maintenance has been an important topic regarding optimizing asset life cycles for a long time. Knowing when an asset is likely to fail and servicing it before the failure occurs reduces downtime. This advancement is expected to save industrial companies $630 billion over the next 15 years.
Currently, with a work order management solution, preventive maintenance can be manually scheduled and work orders will be generated accordingly. These forecasts are often based on manufacturer-recommended service guidelines that consider two factors: time elapsed and meter reading.
AI is going to take maintenance forecasting to a new level. With such large amounts of data available — including maintenance history, recall information and detailed usage rates — AI will be able to identify trends when particular models experience consistent failures. Leveraging this information to suggest preventive maintenance can help to avoid potential downtime. This means AI could access individual assets and recommend specific preventive maintenance.
Overall, AI will help businesses better manage their assets by suggesting maintenance at critical points in the asset’s life cycle. This will extend the asset’s life expectancy, which furthers the return on the original investment. A well-maintained asset will also run more efficiently over time, resulting in more cost-effective energy and resource consumption.
Prepare For The Future
AI’s ability to continuously understand the customer profile and their service requirements can make this tool an integral piece in enhancing the overall customer experience. AI can be used and optimized in such areas as image recognition, work assignment, scheduling and maintenance forecasting to achieve peak business performance. Continuing to apply prescience and autonomy will push the boundaries of the field service experience to new heights.