Artificial Intelligence has increasingly become a key discussion in all industries and its impact in field service management is predicted to be hugely significant, but how should field service organisations leverage this powerful twenty-first-century technology? In the part one of this two-part feature Marne Martin, President of Service Management, IFS outlined why AI in field service is about far more than chatbots, now in the concluding part, she outlines how AI can bring a touch of genius to your field service operations...

Is Artificial Intelligence a key topic for you?! The full white paper on this topic available to fieldservicenews.com subscribers. Click the button below to get fully up to speed!

Sponsored by:

Data usage note: By accessing this content you consent to the contact details submitted when you registered as a subscriber to fieldservicenews.com to be shared with the listed sponsor of this premium content who may contact you for legitimate business reasons to discuss the content of this content.  

Solving Problems When One Isn't Albert Einstein

Human agents are capable of optimally dealing with a customer, and AI can free them up for the most interesting and demanding tasks. In the case of scheduling technicians in the field, humans are just not up to the numerical challenge of adjusting a schedule in an optimal fashion as humans typically focus in on an aspect of a problem to solve rather than finding the best solution overall.

A dynamic scheduling engine (DSE) driven by AI algorithms is designed to solve complex scheduling problems in real time—problems much too complex for any human dispatcher or customer service agent to handle, especially when at times individuals will act myopically based on their area rather than for the greater good of the company and its customers.

"Even a static service schedule can be handled in myriad different ways and decisions regarding which technician to send to which of several jobs in what order are often made based on suboptimal heuristics..."

Even a static service schedule can be handled in myriad different ways and decisions regarding which technician to send to which of several jobs in what order are often made based on suboptimal heuristics.

“Steve’s son is in daycare in this part of town, so I will schedule this appointment last, so he will be close by.” Sometimes jobs are scheduled based on first-in, first scheduled, regardless of the actual urgency of requests that come later.

Manual or traditional software-based scheduling may be a workable solution for service organizations with a very small number of technicians each engaged in a small number of jobs during a day. But it does not take many technicians or jobs for the number of possible solutions to outstrip human computation capabilities either individually or as a group.

Even at the low end of the spectrum, a human dispatcher cannot quickly identify all the possible solutions and pick the best one. With two technicians and four service calls there are already 120 possible solutions— different combinations of technician, job and order. Two technicians, and five service calls yields 720 possible solutions. Four technicians and 10 service calls present a dispatcher with 1,037,836,800 possible solutions.

But the time you get to five technicians that must complete six calls each—a total of 30 calls, you have 12,301,367,000,000,000,000,000, 000,000,000,000,000 possible solutions.

Finding the optimal solution becomes even more complex as additional and rapidly-changing factors are added into the mix:

• Emergent jobs come in that must take precedence over those already scheduled
• SLAs and other contractual requirements demand that some jobs be completed within a given timeframe
• Technician skill sets that influence which tech is sent to which job
• Tools and materials currently in stock on each service vehicle
• The current location of a technician in proximity to each job and to drop locations for inventory that may be required for a job
• The duration of each service call, both in terms of estimated time required to complete the call and whether a current job is running over the estimated time, resulting in knock-on effect on subsequent jobs

Former world chess champion Garry Kasparov, in his book Deep Thinking: Where Machine Intelligence Ends and Human Creativity Begins, makes clear that even his mind is not capable of computing possible solutions and outcomes as rapidly or effectively as an AI algorithm.

"Automating the schedule through AI not only enables a much higher level of service but frees up dispatchers to handle those “beautiful or paradoxical moves” that may delight a customer or solve a tough problem...“

The human mind isn’t a computer; it cannot progress in an orderly fashion down a list of candidate moves and rank them by a score down to the hundredth of a pawn the way a chess machine does,” Kasparov writes. “Even the most disciplined human mind wanders in the heat of competition. This is both a weakness and a strength of human cognition. Sometimes these undisciplined wanderings only weaken your analysis. Other times they lead to inspiration, to beautiful or paradoxical moves that were not on your initial list of candidates.”

Automating the schedule through AI not only enables a much higher level of service but frees up dispatchers to handle those “beautiful or paradoxical moves” that may delight a customer or solve a tough problem.

In the end, collaborating with intelligent machines will get us further faster than going it alone. According to Kasparov, the best chess is now played as grandmasters use computers to analyze positions, opponents’ games and their own games—elevating the level of play. In an interview with the Financial Times, Kasparov, who famously had matches against an early chess supercomputer, described how the best chess is now played by combining “human intuition and understanding of the game of chess with a computer’s brute force of calculation and memory.”

“I introduced what is called advanced chess; human plus machine against another human plus machine,” Kasparov said. “A human plus machine will always beat a super machine. The computer will compensate for our human weaknesses and guarantee we are not making mistakes under pressure … the most important thing is not the strengths of the human player. It is not the power of the computer. But it is the interface. It is the corporation.”

Legacy Approach to Inventory Logistics

Service management for many businesses relies on inventory … if completion of a service call requires inventory and you are out of stock, you cannot meet your commitment to the customer. When a service request cannot be closed on the first visit, it is often because the right part is not on the truck or immediately available.

So, service management software should encompass inventory management functionality, and that functionality should include automated reorder points for each part. The ability to take parts availability into consideration is a critical data set for AI to work on as parts are a critical determinant in first-time fix and job completion where parts are a factor. It also is a key aspect to successful SLA and outcomes-based commercial relationships.

Once inventory data is available and integrated, a powerful DSE may also be configured to influence inventory logistics so parts and materials are housed in warehouses, satellite offices or inventory drop locations closer to anticipated demand, with inventory matched to jobs in a forward or current day schedule. In one very large implementation of IFS Planning and Scheduling™ Optimization—in the London underground transit system—inventory and tools are dropped ahead of each service visit so technicians who ride the subway to the service site can pick them up.

This is only possible with a high degree of coordination between the service schedule, inventory logistics and an AI-driven scheduling tool.

Conclusion

Service organisations should recognise the tremendous potential AI holds—they can harness it to transform their operations, outflank their competitors and disrupt their markets. We are only starting to tap into the different ways AI can be used to better solve the problem of delivering optimal service in a rapidly changing environment as adoption is still lagging despite the real benefits AI brings. The good news is there are several straightforward and easily accessible ways service executives can harness AI technology right now, today.

Is Artificial Intelligence a key topic for you?! The full white paper on this topic available to fieldservicenews.com subscribers. Click the button below to get fully up to speed!

Sponsored by:

Data usage note: By accessing this content you consent to the contact details submitted when you registered as a subscriber to fieldservicenews.com to be shared with the listed sponsor of this premium content who may contact you for legitimate business reasons to discuss the content of this content.