Expert level machine service
FINDIQ offers a knowledge-based assistance system that uses a specially developed AI process to process the expert knowledge of long-standing service technicians/maintainers and uses this to assist inexperienced employees with complex service cases. The aim is to enable everyone to become a service expert so that even more complicated service tasks such as fault diagnosis can be handled quickly and efficiently in the future. To this end, an initial knowledge base is developed with the experts and digitized in a structured manner with little effort. Above all, new knowledge from the field can be easily added and fed back into the knowledge base. The procedure for knowledge digitization and processing is new, unique and optimized for speed of use: 20 hours is required for knowledge digitization on the one hand, 6 clicks for step-by-step instructions on the cause of the fault and service solution. The novel AI process is the core of the solution and, unlike general e.g. NLP approaches, it is adapted to industrial reality. Thus, it is able to provide assistance and immediate results with comparatively little data or knowledge already. Moreover, the computations in the background ensure that these helps are correct and not just “sound semantically good”. It also ensures that the self-learning and self-optimization of the knowledge base remains correct over time and closed via a review system.
FINDIQ GmbH is a young digital company from Ostwestfalen-Lippe, NRW’s industrial region, that specializes in knowledge transfer in machine service. The company is characterized by the development of its own AI processes, which for the first time makes expert knowledge usable in a fast and intelligent way. The primary goal is to increase the service quality and availability of our customers’ machines. Machine manufacturers from small to large, standard to special machine construction will benefit. Industrial companies or service providers also use us when they need or want to provide service themselves and across a broad, heterogeneous machine portfolio.