At Coresystems, we believe that state-of-the-art technology can be used to make work easier and more productive.
Coresystems will use Artificial Intelligence (AI) to make work easier and more productive
In the daily life of a field service technician, there are questions that are asked again and again:
- “What are the common problems you’re facing with this equipment?”
- “What tools should I bring?”
- “Who is the best in my work community to help me complete the task?”
In most field service organizations, these common questions are repeated in communication channels between various staff. Unfortunately, these communication channels are information silos, and are not possible to access in real-time to share time-saving information.
These interactions, repeated day in and day out, are an opportunity for automation and to empower employees to find the information they need in advance and to more quickly resolve customer issue. At Coresystems, we believe that state-of-the-art technology can be used to make work easier and more productive. As an example, we would like to highlight the following use-cases:
Detect the customer intent
Using machine learning to understand customer intent in a chat - such as needing help with or information about a piece of equipment. Proven processes like logistic regression can detect what the customer needs in a written real-time chat. This way, multiple channels can be supported, such as real-time messaging or email.
Find best helper
Using information retrieval processes, the technician is presented with a list of fellow technicians that can help resolve an issue. Similar approaches are used in search engines to provide information in real time. In this case, a technician is presented with a list of people that can help him, ranked by their knowledge and availability.
However, before the service call can be assigned, many constraints must be factored in. For example, the technician must be available during the timeframe of the service call and needs to have the skills to accomplish the job. Other restrictions, such as work time hours, legally mandatory lunch breaks, among others, must also be taken into consideration. In addition, the travel time and distance required to respond to the service call needs to be kept to a minimum.
The increase in constraints and the potential for multiple jobs for multiple resources makes finding an optimal solution for a given problem difficult, which is why a classic Artificial Intelligence (AI) solution is desirable. Since not all possible solutions for a given problem are computationally feasible, the following approaches are taken:
Search algorithms such as tabu search are used that ease the search for possible solutions. Similar approaches have been used by computers to beat human players in chess in recent decades.
Smart scenario pruning
Since not all possible assignment scenarios are feasible, we evaluate only where it is most optimal based on the given knowledge of the scheduling reality.
Using the above mentioned automated approaches, solutions are found more quickly than by humans. However, these solutions aren’t just used for automating the assignment process, but can also be used to help in the planning decision process by providing the dispatcher with tools tailor-made for different planning scenarios that can be configured as needed.