Making Digital Twins a Reality - Waste Water Case Study

Introduction This abstract gives an insight on how the concept of a Digital Twin is being made a reality in the UK water sector. Using a live project as a case study it sets the problem and context, outlines the approach, identifies the status of the project and the next steps. Problems and Objective The water sector has seen significant investment in physical infrastructure in the 30 years since privatisation, to replace and refurbish an aging historical asset base to meet the increasing regulatory and customer expectations. To address future drivers such as Climate change, extreme weather, market de-regulation, increasingly tight quality and service compliance levels, and customer affordability, will require different solutions and decision making capabilities. Concepts such as the 6 Capitals are emerging which place value on more than the physical infrastructure including, financial, manufactured, intellectual, social, human, and natural capitals. The ability to understand the impact of a decision across a number of these values will become essential for the water companies moving forward. It has been recognised by the sector that data is a critical foundation for decision making of the future. The adoption of digital technologies has led to a huge increase in the availability of data, improved integration of the differing data sources is now required to achieve the benefits offered by the disciplines of analytics and data science. Driven by the factors above Atkins was engaged by Thames Water to develop their first Digital Twin for one of the largest sewage treatment assets in Europe. Beckton Sewage Treatment Works covers an area of approximately 250 acres treating the waste of 3.5million customers. The strategic objective of the Beckton Digital Twin (BDT) project is: - To develop a capability that will dynamically inform TW's risk-based investment decision making process to plan for Price Review 24. Additional objectives were also identified to develop the overall goals of the project: - Facilitate operational performance optimisation of the site - Enhance asset maintenance . Approach The concept of Digital Twins is not new. Integration forms a key part of this concept and it is this that underpins our approach. The project uses the Atkins User First, Product Agnostic approach, which is then underpinned by the adoption of the Atkins Asset Management Systems (AMS) model. A digital twin is a transformation project that is enabled by technology. The Atkins Asset Management Systems (AMS) approach, whilst ISO 55000 compliant, also tests the conventional approach to asset management and forces us to look more holistically at the challenge of bringing the technology, people, and process together to identify the interactions between them. It is this User First approach and AMS model that have allowed the team to focus on the user requirements and value without the constraints of a predetermined product. Figure 1 shows the key elements of the AMS model. Project delivery is split into 2 phases: - Discovery - Implementation Discovery phase focused on exploration to determine if the delivery of a Digital Twin is feasible and will deliver value. User exploration The Discovery Phase uses Agile principles including User Stories and Epics to develop requirements. Epic: A large body of work that can be broken down into a number of smaller tasks (user stories) User Story: User stories are short requirements written from the perspective of the user. User exploration is the most important phase of the project as it identifies the objectives and starts to define the functional requirements. We undertook a series of collaborative workshops with the key stakeholders across Thames Water, these were driven using Value Proposition Canvasses designed to draw out the requirements from the different end users. Building on the themes of Pains, Gains, Pain alleviators, and Gain accelerators we collated 270 comments from the stakeholders. The comments were translated into 82 initial User Stories which were consolidated into 13key stories and grouped under the following 4 Epics. Investment Planning - dynamically inform a risk-based investment decision-making process to plan. Operations — understands the performance of the system, in near-real time and identifies any deviations from the optimal or planned and the ability to perform intervention scenarios. Engineering - compares the actual and potential system performance and to determine the optimal process design needed for compliance, whilst balancing the different value capitals the client deems important. Maintenance Strategy- allows an understanding of the health of critical assets to enable predictive and proactive maintenance interventions. Monitors the performance, validation, and updating of the maintenance strategy. The output of the User Exploration is summarised in Figure 2. Technology / Data Exploration A product agnostic reference architecture was developed, and the existing technology stack was mapped against this along with the requirements from the User Exploration. Information Management was assessed against the IAM (Institute of Asset Management) 'Asset Management Maturity Scale & Guidance' The 'As is' maturity was assessed in the range of levels 1 to 2. With a 'To be' IM requirement at level 3. A gap analysis was then used to develop the scope of work for implementation. Figures 3 and 4 summarise the Technology and Data exploration outcomes Benefit Exploration To understand the benefit the DT would bring to Thames Water a comprehensive benefits analysis was undertaken. 6 strategic level benefits were identified, and 36 business level benefits assessed and quantified below these. The benefits were mapped against the user requirements, capabilities, and work packages. This comprehensive mapping gives a line of sight from requirement through to quantified benefits. This enables the development of a delivery programme with a detailed Cost Benefit Analysis (CBA) profile which is powerful when it comes to justifying the investment or understanding the impact of changes to the programme. The CBA identified a value release after 6 months of a potential 18-month implementation programme with a ROI before the end of the project and formed the basis of the business case for implementation. The CBA profile and benefits mapping are summarised in figures 5 and 6. Implementation Phase is in progress and coming to the end of the first phase to deliver a Minimal Viable Product (MVP) focusing on a subset of the user cases and a particular process stream on the site. The Activated Sludge Plant No.2 was selected along with the following user stories from the Operations Epic: - Model event scenarios - Intervention and configuration simulation - Site performance visualisation Current Status The MVP phase has so far successfully delivered: - An integrated site performance dashboard through the integration of key Thames Water systems, including SAP, Site SCADA, Sample Manager and Cockpit. This allows a holistic view of performance in near real time for the first time by consolidating information from previously disparate systems. - 3 Engineering models, Process, Hydraulic and Pneumatic using industry leading products but have developed them further with the vendors to allowing digital integration, increased functionality, and granularity beyond what has been previously achievable. - An integrated model, which integrates the Engineering models together for the first time enabling operational scenarios to be run to determine the outcome on compliance of a particular set of circumstances. These include tide variations, changes to load and the impact of taking assets out of service. The high-level model integration can be seen in Figure 7. - A laser scan of the entire site and 3D representation incorporated into the DT. Next Phase The project is now moving into the next phase which will include: - Using the MVP to make informed decisions and releasing value. - Scaling the MVP across the rest of the effluent treatment streams. - Developing the integrated model to bring together the various effluent treatment streams into a sitewide model. - Expanding the requirements to address the user stories under the Engineering, Investment Planning and Maintenance Epics. Future direction The Beckton Digital Twin is delivering a blueprint for that will be transferable across the UK water market and beyond. The development of the engineering models will benefit the industry providing digitally integrated products with increased functionality and granularity previously unavailable. It also provides a first working digital twin for a Sewage Treatment Works that addresses a wide range of user stories and capabilities in the single solution.