The potential of AI and machine learning in Civil Infrastructure

For decades now, construction companies have been creating huge volumes of digital data.

Where it may have taken physical industries longer than service industries to catch up with the surge of digitisation, digital transformation has already started to happen. There is no lack of digital data in construction organisations— quite the opposite— but that data has never really been consolidated effectively.

This resource will take you on a journey into the future of tech and how AI and machine learning present a whole world of possibilities for the civil infrastructure industry in understanding data.

Technology is the key to construction transformation

As projects have grown and supply chains expanded, civil infrastructure and construction organisations are now facing a host of challenges in delivering projects which makes understanding environments, performance and costs, more and more difficult. We believe new technology will underpin the solution.

Data overload is a very real problem and a barrier to project understanding. Internal silos make interpreting and reporting on data—for the purpose of benchmarking—incredibly difficult and confusing. Outdated legacy systems are burdening teams across a widening supply chain with disjointed data collection, and poor quality data means data visibility has become inadequate for effective collaboration. Visualising data in one powerful platform can provide real-time information and monitoring on infrastructure including condition status, allowing for easier intervention and enhancement planning.

At the same time, in the new world of construction, driving continuous improvement is paramount and the pressure to build more complex projects faster, and more safely, to new quality standards and legislation is mounting. Unless organisations make preparations to become more data-focused and adopt the right technology it will be very hard to maintain the delivery of quality on-time projects.

Culture is key to construction transformation and it starts with three things: data leadership, buy-in and grassroots movements—so making sure data and the platforms used are useful and positively impact the right teams. Unless you’re getting data to people that gives them real value, you will struggle to get a return on tech initiatives. Culture tends to be driven by push and pull factors and when considering the move to a new construction technology ecosystem it’s no different.

Push factors: Pull factors:
The need to work remotelyMakes jobs easier
Financial riskTeams want true collaboration
Increase in project complexityKeeping workers safe & mitigating risk
Siloed dataSustainability targets
Land access issuesCompany policy
Compliance & quality standardsClients want higher technology adoption
Raised client standardsMake real-time decisions to meet deadlines
Public intolerance of industry mistakes

Only a project-centric solution, that can consolidate past data, will enable groups to work together as one to achieve higher-quality and defect-free construction while ensuring on-time delivery.

Sensat’s latest webinar snapshot poll revealed that two-thirds of industry professionals’ greatest push factors for adopting new technology was the need to work remotely, whilst a further quarter said they were challenged by siloed data.

Meanwhile, when asked their greatest pull factors the industry cohort expressed, teams wanted true collaboration (33%), their clients wanted higher technology adoption (33%), and teams wanted to keep workers safe and mitigate risk (22%). These results only further support the need for better data translation and understanding across the asset lifecycle.

Building scaffolding

A single platform used for data understanding

One of the biggest barriers to harnessing data in civil infrastructure is that most of the data sets are sitting in software silos somewhere, which are not accessible in one particular place. Aside from the fact that information is sitting in different repositories and on varying proprietary software, some are not always in digital or computer understandable formats. This makes things more than a little difficult for us to analyse a holistic environment. The good news is that once all of these data sets are represented in one single platform you can begin to master a huge, powerful environment where you can translate and visualise all your site data in one place. We are seeing customers overcoming data pain points and making huge productivity gains by easily accessing, visualising and understanding accurate and up-to-date 3D models of their sites.

At Sensat we call this type of platform a common visualisation environment (or CVE), where you can capture and process the real world in a digital environment. Be it geospatial, topographical, design, logistical or project cost-related data, a common visualisation environment brings all your data sets, people and systems together for greater understanding in a fast, reliable and unified environment your teams can trust.

Only then, once this data is available in a CVE, can we start to execute our vision to make computers understand the real world. Over the next five years, we expect that data and analytics all aggregated in platforms will radically transform both the process and the business of construction, and it is those companies that can embrace data and new technologies that will innovate how we construct things.

Embracing new technologies for quality performance

By delivering all your topographical survey data in one common visualisation environment, your data not only becomes more accessible, but gives you a clear picture of your project all in one place. With tools for analysis, volumetric measurements, earthwork and site decisions, all stakeholders can easily navigate a digital replica of a project and site, understand and see what they need when they want—and all remotely—speeding up everything from the planning and design process, to earthworks, decision-making and where to save on waste.

For example, whereas before, minor changes to the design could mean several weeks or months of backward and forward communications between architects, engineers and owners across several platforms, tools and systems, insights into the effects of changes can now be visible to all almost instantaneously—it’s a game-changer.

Industry professionals have expressed the most valuable benefits from having a common visualisation environment to be improved productivity and collaboration (17%), improved communication(17%), enabled remote working (15%) and recorded and captured stakeholder decisions (15%).

If a CVE can do even half of that mentioned above, imagine the potential when you introduce AI and machine learning into the equation. By adopting AI in construction, we have the potential to make smarter, safer and more sustainable builds across the asset lifecycle by bringing a lot more value to existing data.

Today, the adoption of AI solutions is still quite low in civil infrastructure compared to other industries. AI presents a whole world of possibilities for translating and understanding data to support use cases, but due to lacking data quality from the past, it is difficult for the machine to learn. Data quality needs to improve to get to where we need to be. After all, AI is only as good as the data it’s given. Firms will need significant amounts of quality data to be able to effectively train algorithms. We have a lot to learn but the future is exciting.

As the construction industry continues towards a more data-led, digital future, the adoption of technologies such as a CVE with AI and computer vision will help to establish a lean construction lifecycle and consequently a more sustainable industry.

Check out our eBook where we deep dive into what each of the states looks like today and what they should look like tomorrow.

Further Reading

See Sensat in action

Better data. Accurate pricing. More innovative solutions.

Sensat application screenshot