Multi-domain maturity model for AI and analytic capability in power generation sector: A case study of ABB PAEN Oy
Hummel, Daniel (2021-12-15)
Hummel, Daniel
15.12.2021
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe2021121560746
https://urn.fi/URN:NBN:fi-fe2021121560746
Tiivistelmä
As more smart devices and smart meters are available on the market, industry actors offer AI and analytic suites and platforms where the data streams can be contextualized and leveraged
in pre-made industry specific templates and model, together with self-serving machine learning environments. How can a traditional EPC company, use its domain knowledge in offering
these AI and analytic suites. The assumption made is that there is no inherent value in the AI and analytics suite without data. How should this assumption be incorporated in projects
executed before the operation phase where data from operation is non-existent.This thesis investigate which elements provide a value proposition in the AI and analytic suite and map this
against the domain knowledge of the EPC company. The findings is a novel design in where both operational data is integrated into design for new projects. A survey is also conducted on
the data utilization in the power generation sector based on the same elements. The findings is that while the granularity is low, the quality is good, with an overall maturity between managed and proactive data utilization, which indicate that there are few automated data streams, but that the data is available structurally and in a defined way.
in pre-made industry specific templates and model, together with self-serving machine learning environments. How can a traditional EPC company, use its domain knowledge in offering
these AI and analytic suites. The assumption made is that there is no inherent value in the AI and analytics suite without data. How should this assumption be incorporated in projects
executed before the operation phase where data from operation is non-existent.This thesis investigate which elements provide a value proposition in the AI and analytic suite and map this
against the domain knowledge of the EPC company. The findings is a novel design in where both operational data is integrated into design for new projects. A survey is also conducted on
the data utilization in the power generation sector based on the same elements. The findings is that while the granularity is low, the quality is good, with an overall maturity between managed and proactive data utilization, which indicate that there are few automated data streams, but that the data is available structurally and in a defined way.