August 12, 2020 • 3 min read
How can we use data to improve project delivery efficiency?
The way we collect, store, and use data for capital projects is inefficient. But a lifecycle approach to data can improve decision making both now and in the future.
Many companies operating in the energy sector and other industries struggle to implement digital solutions like big data, artificial intelligence, robotics, and augmented and virtual reality.
Why?
Because we’re bad at sharing data.
How current data sharing creates inefficiencies
Consider how data from a new capital project moves through the stages of a project. It’s handed from stakeholder to stakeholder and converted and adapted by each operating system into which it’s inputted.
Inevitably, some of the detail is lost in translation or is passed across in so many different data sets that it’s not useful unless there is an inordinate spend on a common data model.
The more data created over the life of an asset, the greater the inefficiencies a project is likely to be exposed to. But by changing the way stakeholders interact with each other when it comes to sharing data, this underlying issue can often be addressed – and much more value can be created.
For us, that means a shift in focus from being a contractor to a partner. Becoming more data-centric and considering the full lifecycle of a project’s data, beyond our defined involvement.
The benefits of adopting a lifecycle approach to data
One of the benefits of a data lifecycle approach for capital projects is the potential to reduce the total cost of ownership by running assets as data models.
Each stakeholder’s input contributes to the completion of their aspect of the project and is tracked over time to understand the effect of decisions taken. This builds a picture of how decisions made earlier in the project’s life can affect operation and maintenance costs later.
What’s more, data analysis generated from the concept of a project right through operation, maintenance and decommissioning can create better visibility and ensure a highly reliable, safe, and sustainable asset.
Imagine if you collated all the data on project reworks, then brought together the project team and data analysts to identify the root cause. This could have a substantial impact, not only on the project at hand, but for upcoming projects, too. The team could use this data to better model asset integrity and reliability and the change over an asset’s lifetime.
The data could also be used to model how the asset is influenced by factors such as temperature, corrosion and operating hours, and help operators understand the causes of failures and unexpected events.
Physical data can be digitized too
Lifecycle data does not need to be numeric. Data from images and videos are increasingly collated and analyzed as a source of intelligence.
For example, video data can be used to track the movement of staff on site and match work orders to workers and confirm whether all the right paperwork is in place for the work to proceed.
We can use optical gas imaging to detect gas leaks, helping to identify and prioritize repairs and reduce greenhouse gas emissions. Video feeds of flares, coke drum operations and internal inspections can, when married with process data, lead to greater reliability over time.
Next steps to implementation
Using data as a central theme to lower energy intensity will go hand in hand with new and upgraded digital plants. This will lead the industry to move away from a transactional approach to capital projects, to work more collaboratively across the project’s lifecycle.
To get the most out of the opportunity, the industry must focus on building ‘sensored’ assets which feed into a common data model. This allows us to collect data from the sensor for the lifecycle of the asset.
Moving forward, we won’t need pure data analysts, process data analysts and reliability data analysts. Instead, we’ll need hybrids and engineers who can work with data and blend it with their own experience, to produce the most compelling insights.
Other industries such as finance, automotive, and fast moving consumer goods have already reaped the benefits of a lifecycle approach to data. It’s time we did so too.
