Decision Lab Case Study with SIEMENS
Requirement
The Siemens Industrial turbine division needed a product that would be a digital twin representation of their portfolio of industrial gas turbines, that allows scenarios to be modelled as well as the operations and maintenance activities of the fleet to be optimised.
Challenge
Siemens’s industrial turbine division operates in a complex and dynamic environment, given its multiple clients across the globe and the inherent unpredictability of asset management. Furthermore, modelling the complex lifecycle operations of industrial gas turbines as well as the interactions of Siemens’ asset network was a real challenge, one that was exacerbated by the need for a tool that could manage the version control of the large input datasets.
Solution
Decision Lab developed the ATOM tool and the corresponding version control management web app that allows Siemens to optimise maintenance and mitigate risks making sense of the available data, improving their decision-making process while offering a great user experience.
Optimising the maintenance activities of Siemens’ portfolio of gas turbines (Requirement)
Siemens has a large portfolio of industrial gas turbines for a variety of clients in the oil & gas and power generation sectors, the management of which requires the use of vast quantities of data. A new approach that would replace the outdated models that Siemens was using to address the issue was necessary, one that utilises the abundance of available data to assist them in making more informed decisions.
Forecasting highly complex operations (Challenges)
Our task was to design and build a digital twin solution and develop a flexible simulation model that enables the user to carry out what-if analysis and thus improve the efficiency of the operations. The management of Siemens’s large fleet of assets is a complex problem, given the number of interactions in the network, the need to track and monitor assets across multiple locations spanning different continents, the various operational constraints as well as the need to minimise downtime for the clients. Furthermore, in a later phase of the model, it became apparent that the management of the version control of the input data needed a new, bespoke solution to make the whole process less tedious, more robust as well as a better user experience.
Squeezing vast data into a digital twin of a global-scale operation (Solution)
We started by designing a solution that would essentially be a digital representation of Siemens’ assets operations. It uses the vast quantities of data available to integrate customers, supply chain, production, and maintenance to improve productivity and efficiency in customer operations and asset management. At its core ATOM achieves this by modelling the detailed intricacies of customer operations, maintenance facility operations, engine characteristics, and supply-chain logistics across the whole fleet and the operational cycle, resulting in an advanced global-scale maintenance model for Siemens.
A positive simulation – just like the real thing (How do we achieve this?)
The agent-based simulation model we developed captures the client’s complex business process to identify removals, failures, and bottlenecks and run what-if scenarios. The capability has been transitioned as an enterprise solution on the Siemens network, while recently a web app that acts as a version control management tool of the input datasets was developed, which can facilitate the queuing of the jobs created by users and log any errors raised as well as compute run-time statistics. The resulting web app greatly enhances the model’s capabilities, making the process more robust as well as significantly improving the user experience.
ATOM – a key building block to Siemen’s successful business decisions in maintenance (Future)
With the introduction of ATOM, Siemens was able to gain a better understanding of the requirements for spares and likely maintenance downtime. This, in turn, allowed them to mitigate risks to customers by optimising maintenance lines and placing the right spares in the right locations at the appropriate time. The web app built by Decision Lab in a later phase can be used to manage version control of the model input datasets, including uploading new versions and modifying existing datasets, making the whole process more robust. The app then queues jobs to run on the AnyLogic Private Cloud (ALPC) before the results can be obtained to be processed by Siemens. ATOM has led to better business decisions from Siemens and maintenance and ongoing support contract with Decision Lab.