LLM-supported analysis of tickets in field service management
Fieldcode is a cloud-based web application for managing field service assignments. Fieldcode GmbH, which is based in Nuremberg and operates internationally, mainly serves IT outsourcing companies and manufacturers that use the company's platform to coordinate field service assignments. As part of the joint AI pilot project, an LLM (Large Language Model) system is being developed to optimise ticket diagnostics. This system will automatically provide recommendations as to whether a problem can be solved remotely or whether a field service call and certain spare parts are required.
Web application for managing field service assignments from the Franconia region
Fieldcode is a leading field service management software based on 20 years of global experience. It offers a zero-touch process that automates the handling of tickets from creation to technicians without manual intervention and facilitates the work of dispatchers. As the most cost-efficient solution on the market, Fieldcode offers flexible per-use pricing and advanced BI forecasting . It helps organisations manage the entire lifecycle of their field service, ensuring optimal efficiency and improved customer satisfaction.
Challenge: Optimisation of ticket diagnostics
The company takes care to plan and carry out field service assignments as efficiently and successfully as possible, with the first fix rate (successful resolution of the problem in the first assignment, no further visits necessary) and the suitability of the selected spare parts being of particular importance. The correct elaboration of the instructions and spare parts during the processing of the ticket is essential in order to be able to successfully solve the problem on the first visit and thus avoid further visits. The selection of the required spare parts in particular is very difficult, which is why more spare parts are often ordered than are actually required. In addition, spare parts are sometimes ordered and technicians dispatched even though the problem can also be solved remotely. Fieldcode estimates that half of all spare parts ordered and dispatched are not needed. A large proportion of these unused spare parts are returned and reconditioned, while the rest are disposed of.
Process optimisation with the help of AI
The AI pilot project with the Green-AI Hub Mittelstand optimises the ticket diagnostics process through the use of Large language models (LLMs). Artificial intelligence accesses information much faster and more comprehensively than clerks, and LLMs are particularly good at understanding and processing language. These LLMs can therefore be used to analyse ticket descriptions.
In addition, the language model must be enriched with specific expert knowledge, and the Retrieval Augmented Generation (RAG) process is integrated to implement this in a resource-efficient manner. Supplementary information, such as historical tickets and instructions for use, is stored in a database and retrieved as needed.
Process optimisation through AI-based analysis of tickets in field service management
The optimisation of processes in the areas of transport and service enables a more efficient and resource-saving way of working. Employees are supported by the AI system when analysing tickets so that the problem can be solved in a more targeted manner. The improvements lead to fewer journeys by technicians to customers, as the first-fix rate is increased and possible remote processing is better recognised and implemented.
Saving resources through the use of artificial intelligence
The AI solution means that fewer visits to customers are necessary, resulting in less fuel consumption . This more targeted selection of spare parts reduces the number of spare parts shipped. This, in turn, means that fewer parts need to be reconditioned or disposed of, and transport damage occurs less frequently.
Presentation of the pilotproject with Fieldcode
- Alexander Schmid, Fieldcode
- Fabian Reichwald, Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI)
Technology
AI capability: analysing
AI models: Large Language Models (LLM), Retrieval-Augmented Generation (RAG)
Value creation
Phase: service
Aim of AI: optimisation through more precise analysis of tickets and derivation of relevant information
Resource efficiency
Saving CO2, as fewer journeys to customers are necessary and of spare parts and materials thanks to a better selection.
Company
Industry: Field service management (FSM) Software / Software-as-a-Service (SaaS)