In every company, there are employees who have built up knowledge through experience over the years. This implicit knowledge, for example in the form of practical routines, personal experience or tricks of the trade, is acquired individually in everyday working life, but is rarely recorded in writing as it is difficult to put into words. It becomes critical when these experienced employees leave the company. This is because a large part of their undocumented experience often disappears with them.
The situation is different with explicit knowledge: Information that is already available in written form, such as in emails or minutes, is comparatively easy to transfer to a digital knowledge management system. This means that other employees can also access and benefit from it.
Grid operators face a particular challenge here: demographic change and changes to the electricity grid due to renewable energies are increasing maintenance costs, while valuable expertise is increasingly being lost due to staff turnover. This is where digitally supported knowledge management can provide crucial help and counteract the challenges.
A look at practice: when knowledge should not be lost
Employee A has been with the company for 40 years and is about to retire. During his time as a maintenance technician, he has serviced every grid station and switch cabinet in his area. For example, he knows exactly that control cabinet A needs more solvent and that there is no mobile phone reception for the switching call in network station B. He can prepare himself accordingly. When he leaves the company, he wants to pass on his experience to his colleagues. As there are many similar control cabinets in his network area, he finds it difficult to assign the experience values to the respective control cabinets.
From empirical knowledge to an AI-supported assistance system
This is where the method from the AProSys research project comes in. The aim is to establish comprehensive knowledge management in the energy sector. Both explicit and implicit knowledge is to be systematically collected, structured and utilised for an AI-based assistance system.
The method uses geodata from grid stations, company information and the various maintenance sub-processes to record knowledge in an organised manner. This approach combines knowledge management with AI assistance systems. The project thus provides initial insights into the development of current, domain-specific assistance systems that support companies in their operations with up-to-date information.
Results presented at ECIS 2025
The findings from the project were documented in the paper "Igniting Knowledge Management for Assistance Systems in Maintenance: A Method for Knowledge Gathering" and presented at the European Conference on Information Systems (ECIS) 2025 in Amman, Jordan, by Alexander Skolik, research associate in business informatics, in particular business information systems.
The conference was organised as a hybrid event and was held under the motto "Creating shared value for an intelligent future".
Dr Philipp zur Heiden, Tamino Donner and Jennifer Priefer were also involved.
Paper:https://aisel.aisnet.org/ecis2025/ent_system/ent_system/2/
