Washington River Protection Solutions (WRPS) manages clean-up of the Hanford site, a 586-square-mile area owned by the US government which houses 56 million gallons of nuclear waste.

Since 2016, Royal HaskoningDHV has worked as part of the extended WRPS Mission Analysis Engineering Team to introduce digital twins and predictive simulation to the project.

In the first phase, WRPS created a digital twin ecosystem which uses WITNESS Horizon predictive simulation software to identify bottlenecks and direct investment more efficiently.

The second phase of collaboration has focused on evolving WRPS’ digital twin capabilities to facilitate better, faster decision making, finding new ways to accelerate project completion.

Project facts

Challenge Finding new ways to speed up data analysis for informed decisions
Client Washington River Protection Solutions
Location Washington state, USA
Solution An app that uses cloud-based simulation, artificial intelligence and machine learning to provide data-driven answers to complex queries.


cloud-computing-and-ai-in-digital-twins | Royal HaskoningDHVcloud-computing-and-ai-in-digital-twins | Royal HaskoningDHVcloud-computing-and-ai-in-digital-twins | Royal HaskoningDHV
Use AI and machine learning to automate data analysisReduce compute time to speed up analysis and boost efficiencyIncrease flexibility and enhance business continuity


After the success of the initial phase of the digital twin project, stakeholders within WRPS and the Department of Energy were increasingly impressed with predictive simulation – and that meant there was rising demand for modelling to be integrated into decision-making processes.

With such a delicate and technical operation to manage and analyse, the team needed to answer extremely complex questions. A single query about improving throughput could easily involve as many as 80 separate scenarios, all of which needed to be run and assessed for their potential impact on the project. 

It took hours to prepare and check a vast number of input files, days of compute time to run scenarios, and yet more days to analyse the large data volumes. By the time this process was complete, it could easily take a month to get through just one set of experiments – slowing decision-making and potentially adding more delays to the project.

So, the next step for the WRPS Mission Analysis Engineering Team and Royal HaskoningDHV was to look for ways to introduce more automation, artificial intelligence (AI) and machine learning into its simulation processes, and get actionable data more quickly.


The WRPS Mission Analysis Engineering Team and Royal HaskoningDHV focused on two ways to boost modelling efficiency: reducing compute time and speeding up analysis, which led to a two-pronged approach:

  • - Developing a cloud-based experimentation app, to provide more power and flexibility for what -if scenarios.
  • - Using AI and machine learning to automate data analysis
Using, a Royal HaskoningDHV web service, the team built a bespoke app for mass experimentation. Using the cloud drastically reduces overall compute time, as it provides access to dedicated, scalable simulation cores, enabling the team to run multiple scenarios simultaneously. It also boosts flexibility and enhances business continuity.
The team is using AI and machine learning to analyse simulation outcomes, speed up individual experiments, and identify possible improvements via automatic bottleneck detection.
This accelerates the iterative experimentation process, helping home in on blind spots and answer questions more quickly.
For more information on how the app works, take a look at our infographic.


Increasing its use of predictive simulation has helped WRPS get data-driven answers to its most complex questions. And thanks to cloud computing, AI and machine learning, the team can now deliver those answers on demand.

It used to take up to 25 working days to answer a standard 80-scenario question, but now the process can be completed in as little as two days. This has fundamentally changed how the team uses its digital twins, boosting efficiency and reducing human error throughout the process.

Let’s look at one key example of this new approach in action. A major part of the clean-up process involves retrieving tanks containing processed waste. Using the cloud app to access its digital twin ecosystem, the team was able to run hundreds of scenarios and answer critical questions about tank volumes, timescales and safe retrieval processes in less than two weeks – a process that would have previously taken days’ worth of cutting and pasting data into spreadsheets, followed by weeks of data analysis. In waste management, there are hefty fines for missing deadlines, and the quick, detailed analysis helped facilitate safe, on-time retrieval.

As an added benefit, the ability to work in the cloud has been crucial to keeping the project running during the COVID-19 pandemic, as the team can access the experimentation service remotely.

“The modelling is changing hearts and minds. Our achievements are testament to the strong collaboration internally and with Royal HaskoningDHV. The Department of Energy has even asked us to lead training sessions – using us as an example of how predictive simulation should inform decision making.”
WRPS team member

Want to discuss whether your organisation is ready

for a digital twin?

Contact us

Keeping up to date with the latest Digital Twin news?

We've got you covered