Predictive Digital Twins for Offshore Wind Fabrication

There are some wild renewable energy ambitions out there. First, the EU wants to build 400GW of offshore wind by 2050. Not to be out done, the UK wants to build 50GW by just 2030. But this pales in comparison to Fortescue Future Industries (FFI) who wants to build 450GW by 2030. I’ll leave it to you to decide whether a billionaire can build more renewable energy in 8 years than the entire EU can in 30 years. As I described here, offshore wind is key to all these ambitions. The question is, how will the offshore wind sector scale up to deliver them?

In my previous article I explained how our Predictive Digital Twins had helped people in other growth sectors scale their production facilities or meet increasingly demanding targets from existing facilities. Here I’ll take you through two examples directly from offshore wind development.

Delivering two contracts from one site

A fixed bottom offshore wind turbine normally consists of monopile or sometimes suction bucket foundations on the seabed. On top of this is a 50-80m tall steel frame structure called a jacket and a single offshore wind farm could have a hundred of these. When one of our clients (that has asked to remain anonymous) won two contracts (one for 54 and one for 32 jackets) in two different parts of the world, they turned to our Predictive Digital Twin simulations to help them understand how they could deliver these from one fabrication site.

The fabrication project manager told us that they’d tried to use Primavera (critical path planning software) to create the multiple scheduling scenarios, but it became so complicated they gave up. This gave them a good grounding in the steps of their process which is an important point to start for any simulation.

Using these process steps, we created a Discrete Event Simulation model of the fabrication. This had inputs covering contracts (dates, quantities, and vessel schedules), work content (steps, tasks, and jobs database), layout (workstations, configurations, stocks, and logistics), resources (both HR and equipment) and a Bill of Materials with up to 10 layers of information. But because this process is dependent on people, we also included shifts, holidays and working conditions.

Our model was then used with multiple scenarios of throughput, process configuration, site layout, and storage options. The outputs were visualised in a Business Intelligence Dashboard including estimated schedules, workstation utilization, stock level, storage requirements, and vessel waiting times.

This allowed the team to evaluate scenarios to understand how to best utilize their site to meet two contracts. Our Predictive Digital Twins was that successful that it’s now been used across multiple sites. Enabling optimal jacket fabrication to support offshore wind growth.

To bid or not to bid

On the other hand, companies often want to understand if it’s worth investing in expansion of their site to bid for contracts in the growing offshore wind sector. This was the question that the Institut Supérieur d’Etudes Logistiques (ISEL) in Le Havre explored using our Predictive Digital Twin software in a research project. With the aim of advising whether Fouré Lagadec should invest to bid for a contract to fabricate offshore wind turbine towers.

As shown in the image below from the ISEL team, first they identified the question to be asked. They collected and analysed existing data, assessed it for adequacy, and cleaned it up. In parallel they configured our Predictive Digital Twin software to the specifics of the question and developed a model based on the fabrication process. With this they could run multiple scenarios using the real data from the site to make an assessment.

Similarly to the jacket foundation site, Fouré Lagadec were interested in how many workstations they’d need, the level of resources they’d need to commit to and how much of the site would be needed to store work in progress. The model covered every stage from receipt of parts to the site, manufacturing of the section and on-site storage before being loaded on to vessels.

The simulation team then estimated the investment needed and the process configuration to grow from 3 to 50, 70 or 100 masts. Including whether an increase in shift patterns would support the business case.

Ultimately the simulation team found that in this case the investment needed to meet the contract at that site wouldn’t be matched by the size of the contract, potentially saving Fouré Lagadec from investing millions that wouldn’t create the return needed for their business.

Dan Smith is the Digital Renewable Energy Commercial Lead at Royal HaskoningDHV. He helps companies across the renewable energy supply chain solve problems using leading Digital Twin technology from Lanner and Royal HaskoningDHV. To learn more and connect with Dan, find him on LinkedIn here.  

Ready to learn more? Click here to schedule a call with Dan to discuss your needs and the renewable energy solutions offered by Royal HaskoningDHV.