Refine
Document Type
Language
- English (3)
Is part of the Bibliography
- no (3)
Keywords
- Data Analytics (3) (remove)
Institute
Electricity generated by wind turbines (WT) is a pillar of the transition to renewable energy [1]. In order to economically utilize WTs, operating and maintenance costs, which account for 25% of total electricity generation costs in onshore WTs, are a focus of cost reduction activities [2]. A prescriptive maintenance approach can support in achieving this goal. Prescriptive maintenance is a maintenance approach, where asset condition data is collected and analyzed to recommend specific actions to prevent breakdowns and reduce downtimes. However, the processing and analysis of data is quite complex. Especially unstructured data (such as comments of service technicians in free text fields) is often left unused, as companies, mostly SMEs lack the capacity to carry out these analyses. In this work we propose an approach to utilize the information from service reports, maintenance reports as well as status records from SCADA systems for the development of a prescriptive maintenance approach to onshore WTs. To achieve this, an ontology was utilized in this approach to codify implicit knowledge of service technicians and aid in making unstructured data usable for further analysis. The ontology was used to link historical service and maintenance reports with status codes, thus enabling automated analysis. In interviews with WT topic experts and through further research, damage mechanisms and corresponding maintenance measures were identified and a measure catalogue was developed to support service and maintenance activities. The recognition of the root cause of problems allows for a prescriptive maintenance approach that recommends targeted actions to reduce downtimes and optimize maintenance activities, it also allows to effectively control the outcome of maintenance activities and optimize their execution.
Industry 4.0 and the consequent necessity of digitalization has also impli-cations to the field of procurement, resulting in the so-called term of Procurement 4.0. Digitalization can be a valuable tool to increase the efficiency of the procurement organization and to exploit new opportunities of growth. A mandatory requirement to perform the digital transformation is an increased transparency along the procurement process chain. This paper aims to conceptualize a digital shadow for the procurement process in manufacturing industry as a basis for advanced data analytics procedures. The term digital shadow stands for a sufficiently accurate, digital image of a compa-ny's processes, information and data. This image is needed to create a real-time eval-uable basis of all relevant data in order to finally derive recommendations for action. The formation of the Digital Shadow is thus a central field of action for Industrie 4.0 and forms the basis for all further activities.
The acquisition, processing and analysis of internal and external data is one of the key competitive factors for corporate innovation and competitive advantage. Many firms invest a significant amount of resources to take advantage of advanced analytics methods. Machine learning methods are used to identify patterns in structured and unstructured data and increase predictive capabilities. The related methods are of particular interest when previously undiscovered and unknown structures are discovered in comprehensive data sets in order to more accurately predict the outcome of manufacturing or production processes based on a multitude of parameter settings. So far, this knowledge is often part of the individual or collective knowledge of experts and expert teams, but rarely explicit and therefore not replicable for future applications. On the one hand, it is demonstrated in this paper how different machine learning algorithms have been applied to better predict the output quality in the process industry. On the other hand, it is explained how the application of machine learning methods could contribute to making previously not accessible process knowledge explicit. In order to increase the prognostic accuracy of the model diferrent methods were combined, later on compared and evaluated within an industrial case. In this paper a comprehensive approach to knowledge-based process engineering is being presented.