← Back to Resources

Smart turbine spare parts management at Siemens Energy

CASE STUDY - ENGINEERING & MANUFACTURING

Siemens Energy application engineers leveraged metaphactory's low-code approach to build a smart turbine spare parts application that allows turbine service engineers to save thousands of hours on manual effort.

 

Gas turbines of Siemens Energy are used worldwide in different environments and with customer specific configurations. Managing a broad variety of spare parts and configurations for each turbine is a challenge. metaphactory and Amazon Neptune enabled Siemens Energy to build a Turbine Knowledge Graph and visualize the connections between similar parts across the entire fleet of large gas turbines.

The customer

With its products, solutions, systems, and services, Siemens Energy addresses the extraction, processing, and transport of oil and gas as well as power and heat generation in central and distributed thermal power plants, and power transmission and technologies for the energy transformation, including storage and sector- coupling solutions. The Generation Service business unit of Siemens Energy is responsible for global services and maintenance around large gas and steam turbines and generators.

The challenge

Siemens Energy large gas turbines are used in power plants around the world in different environments, be it with the ultimate goal of power generation for households or for industrial complexes.

 

The turbines generally require customer-specific configurations, fitting the customers' environment and line of business. This results in heterogenous digital representations of turbine configurations. Lacking a standardized way for modeling machine structures, a broad variety of tools was in use for documenting these heterogeneous representations.

 

When large gas turbines are scheduled for maintenance or repairs, the correct spare parts need to be provisioned in the right amount and at the right time in order to avoid extended turbine downtimes. However, because these turbines are configured to fit the customer environment, each customer needs an individual spare part catalog and packages for maintenance. Since turbine configurations were maintained in heterogenous tools and formats, high manual effort was associated with creating and maintaining customer-specific digital machine structures.

 

To address these challenges, the Generation Service business unit of Siemens Energy put together a team dedicated to creating and perfecting a smart and targeted solution for maintaining spare parts of large gas turbines.

The solution

Siemens Energy decided to build a software application that would provide business users, i.e., turbine service engineers, with a unified overview of all large gas turbine configurations and spare parts catalogs previously stored over multiple, diverse data sources. This software application would allow Siemens Energy turbine service engineers to compare existing configurations, identify identical parts, and give recommendations on the quantities for required spare parts, thus reducing warehouse inventory and avoiding extended turbine downtimes. For this, Siemens Energy opted to leverage the power of Knowledge Graphs and develop a solution comprising metaphactory and Amazon Neptune.

 

Knowledge Graphs – A Unified View over Your Data

 

When configuring a new turbine for a customer, turbine service engineers generate a machine structure documentation. However, this was previously done in heterogenous formats and custom documentations often listed spare parts in differing sections. This also resulted in identical spare parts often being recorded multiple times in the internal systems.

 

The data represented in a turbine documentation is hierarchical, which means that the structure of the documentation can be very well represented as a graph.

 

Excerpt of the Knowledge Graph showing a request item and connected resources

Excerpt of the Knowledge Graph showing a request item and connected resources

 

The flexibility provided by Knowledge Graphs to introduce new links and connections between spare parts and enable insights into similarities between machine structures despite them being documented differently was unprecedented. "Already first experiments with the RDF [Resource Description Framework] graph data model and graph databases were very successful and we could really see this working not only for this use case, but for many others in the future," said Lutz Lukas, IT Solution Architect at Siemens Energy.

 

Visual exploration of spare parts

Visual exploration of spare parts

 

"The Siemens Energy solution engineers managed to truly tap the potential of Knowledge Graphs and show what is possible, laying the stepping stone for further projects that leverage semantic and knowledge graph technologies," said Dr. Daniel Herzig-Sommer, COO at metaphacts.

 

Amazon Neptune, a managed graph database service, fits perfectly into the cloud-first strategy driven by Siemens Energy IT, which focuses on reliability, scalability, reduction of maintenance and integration with their existing platform on Amazon Web Services (AWS). metaphactory for Amazon Neptune was purchased via the AWS Marketplace, which allowed Siemens to directly get metaphactory running in their private AWS cloud and connected to their Amazon Neptune service.

 

Agile Development of End-user Applications

 

metaphactory enabled the Generation Service business unit of Siemens Energy to build a Knowledge Graph on top of Amazon Neptune. The Knowledge Graph provides a unified view of the data coming from multiple, diverse data sources and allows business users to explore and analyze gas turbine configurations from different perspectives.

 

The goal was to bring data to business users as fast as possible and ensure that the final product matches business requirements. metaphactory is a low-code platform used to rapidly build data-driven, end-user facing applications matching individual needs. The involved Siemens Energy solution engineers used the metaphactory platform and its out-of-the-box components to rapidly develop a custom application on top of Amazon Neptune. The development of the application was done in-house with metaphacts providing support for metaphactory deployment and configuration questions.

 

metaphactory allowed business users to already use the application for real world tasks during the development process. As such, feedback was collected early on and could be implemented in an agile way, allowing for new features to be added incrementally along the way. "metaphactory offers many out-of-the-box components and allowed us a shorter time to market. The fact that we could prototype and develop our application in-house without much coding convinced us," said Amit Vaidya, IT Project Lead at Siemens Energy.

 

Data Curation and Data Quality Assurance

 

The data quality workbench delivered with metaphactory allowed Siemens Energy to find inconsistencies, curate the data and constantly monitor data quality. Identified data quality issues can now be traced back to the data source and resolved at their origin.

 

Fleet-wide Search and Visualization

 

metaphactory's search interpretation engine leverages advanced algorithms to recognize the data structure and schema and return relevant results within seconds. In this case, it uses the data model to support business users in building targeted, natural language queries to quickly find spare parts of large gas turbines and analyze where these spare parts are in use. And all this without the user having to know the data model and the relations between concepts.

 

Intuitive end-user search interface across the fleet of large gas turbines

Intuitive end-user search interface across the fleet of large gas turbiness

 

From there, users can explore further using metaphactory's rich set of components for interactive visualization and exploration and gain meaningful insights into relations between spare parts, turbines, customer-specific configurations, maintenance schedules, and maintenance history.

The results

"A key differentiator of the metaphactory platform was that it delivers a great combination of graph data management, custom visualizations, data quality assurance, and natural language keyword search in one platform," said Amit Vaidya, IT Project Lead at Siemens Energy. metaphactory allowed business users to test and interact with the data and provide feedback while the data experts were building the data model and integrating data from various sources into the Knowledge Graph. The fast development of the spare parts management application resulted in shorter time to market for the internal business solution.

 

Data quality assurance is another key aspect which was critical to the business solution. metaphactory alongside the in-house developed application helps business users identify and correct inconsistencies in gas turbine documentation, thus reducing the number of error rectification requests and delivering an intuitive user experience.

 

Access to the right data through the intelligent keyword search framework resulted in higher productivity and an intuitive experience when exploring the Knowledge Graph. The previous approach, which implied asking a different department for a custom SAP report, was replaced by an on-demand report generation engine which not only reveals more insights into the data but also resulted in time savings of up to 1,500 hours in the first year already.

 

"Through fleet-wide analysis our business users can take data-driven decisions when optimizing outage- specific spare part packages. An optimized package recommendation means happy customers, as it allows us to give them more comprehensive and precise recommendations as to which spare parts to order for which outage," concluded Lutz Lukas, IT Solution Architect at Siemens Energy.