Reuse infrastructure project data to leverage AI and machine learning

Information is essential to the successful implementation of technologies such as AI and machine learning, but too often the data in infrastructure projects is not reused. Bentley Systems’ Claire Lutkowski Speaks

Our world runs on data. The advent of BIM and the proliferation of IoT sensing devices has created a veritable data flood. we are drowning in it. But surprisingly we don’t use it.

A report from consulting firm FMI Corporation specifically states that 96% of infrastructure project data is never reused.

Some might argue that certain pieces of data are project-specific and should not be reused, but certainly not 96% of the time.

One of the reasons why infrastructure project data cannot be leveraged is that it occurs whenever deliverables are passed from one company to another, or from one phase of the asset lifecycle to another (e.g. project design and delivery-to-construction, construction-to-construction handover, etc.). during commissioning), the data will be lost.

90% of all data is unstructured data

Perhaps the tools and systems used in each phase are not communicating with each other, or the data being produced in one phase is incompatible with the structure of the data in the next phase.

As a result, a lot of the hard work you did in one phase is lost, and in many ways you start over with less fidelity and less detail.

The same FMI study reported that 90% of the total data FMI accounts generate during engineering and construction is unstructured.

Unstructured data can be difficult, if not impossible, to convert and migrate from one tool to another. This is one of the main causes of information lost along the way.

The result is a lack of data visibility and communication and collaboration challenges as stakeholders do not see or talk to the same data set.

It also means that decision makers are making decisions based on incomplete data because some of the data was not transferred. And siled data is left behind, accelerating obsolescence.

Leverage AI and machine learning to deliver sustainable outcomes

For infrastructure departments to take full advantage of tools such as artificial intelligence and machine learning, stakeholders need robust datasets containing structured and unstructured data that can be mined for patterns and insights. is.

AEC firms and asset managers need consistent and complete data sets to facilitate communication and collaboration.

A consistent and complete data set ensures that the complete carbon footprint of an asset can be measured from project implementation through construction, operation and potential decommissioning.

This helps the sector drive sustainable outcomes.

Robust and accurate data that can be harnessed by machine learning and artificial intelligence tools can also help ensure the accuracy of design intent, improve efficiencies throughout the asset lifecycle, and provide insights otherwise unobtainable. Helpful.

How can the infrastructure sector achieve this?

From a technology perspective, creating and maintaining rich datasets requires an open platform that can work with multiple tools.

Interoperability between these tools is also required, and data must be able to be shared securely and gracefully across the supply chain.

You also need to acquire and maintain data. Data requirements should be standard across the lifecycle.

We use P&IDs to tie things together, but let’s look deeper.

Taking data definition further up the lifecycle and defining what is needed at each phase ensures that you get the right data as your project progresses.

Utilization of digital twin function

Using digital twin capabilities, infrastructure stakeholders can combine and integrate data from disparate sources to create geospatially referenced data about their assets such as carbon footprint, emissions, and as-designed data. You can create a set of data silos in a single cohesive bank of data points across As-built data, maintenance information, or asset performance information.

But even digital twins can become siled and stagnant if not done right.

To unlock true value across the entire infrastructure lifecycle, digital twin technology must have an open foundation for building on infrastructure schemas with standards for transforming data from one application to another. Must have.

Only then can the digital twin be leveraged as a dynamic, comprehensive and living dataset.

Only then can you uncover all the dark data currently missing across the lifecycle of your infrastructure and assets to gain true insight.

Is the job worth the effort? Absolutely.

Rich datasets provide actionable insights to improve project decision-making and operational performance.

These insights can shed light on previously undetected trends and enable organizations to proactively correct course, helping design firms, EPCs, builders, asset owners and operators, and the public at large. and so on, leading to better outcomes for all stakeholders.

it’s all about data

Stakeholders can leverage data-rich machine learning, artificial intelligence, and other tools to gather data insights for carbon accounting, predictive maintenance recommendations, and better sustainable outcomes. increase.

But it all comes down to data.

Don’t settle for 4% data reuse. There is a 96% chance of a smarter infrastructure outcome across the lifecycle.

With infrastructure project data as the foundation, it can be collaborated, leveraged, and reused to be more efficient and insightful.

Given the increasing demand for infrastructure projects, growing backlog of orders, resource scarcity and the urgent need to reduce climate change, we need to get the basics right. I can’t help it.

Claire Latkowski