Process Digital Twins

What are Process Digital Twins?

Digital twins are a virtual representation that serves as the real-time digital counterpart of a physical object or process that spans its lifecycle. Process Digital Twins use real-time data, simulation, machine learning, and also reasoning to help with decision-making.

There are three distinct parts to a digital twin system:

  • the physical product or equipment
  • the digital or virtual representation
  • and also, the connections between the two

    In an automation and control context, we use Process Digital Twins for monitoring, diagnostics, and prognostics to optimise asset performance and utilisation. In effect, they help to find the cause of issues and improve productivity.

    The prognosis that a process digital twin provides improves by consolidating additional sensory data, human expertise, physical model simulation, and machine learning.

    Our Process Digital Twin Journey

    Our expertise focuses on real-time processes directly associated with producing ‘a product’ within the environment commonly called the ‘plant floor’. We also work with others in expanded scopes that need this data. Common applications are found in mining, manufacturing, and infrastructure services.

    Our team uses process digital twins as part of the delivery of its business improvement initiatives. This technology has proven to simplify processes and generate improved outcomes.

    To maximise the effectiveness of a Process Digital Twin, it is vital that you implement a structured approach. The steps are summarised below;

    DIGITAL TWIN PROCESS JOURNEY

    Objective

    Firstly, consider objectives which may include specific targets, the amendment of existing issues, and sought outcomes.

    Value and metric

    Secondly, determine the value of the objectives. In doing so, categories can include efficiency, production output, and reduced downtime.

    Approach

    Thirdly, prepare the plan or approach for achieving the objectives, including tools and resources with assessments for suitability.

    Data access

    Fourthly, source and access live and historical information. Often, you will convert raw data to more insightful information using calculations and less complex analytics.

    Digital twin format

    Fifthly, consider the objectives and outcomes desired to determine the digital twin format. As such, this includes visualisations, simulations, machine learning, and artificial intelligence as required.

    Analytics

    Lastly and most importantly, outcomes will improve as analytics become more sophisticated. Furthermore, any iterations from outputs will continue to improve optimisation.