<< Back to Home

Why AI-based personalized predictions are the future of construction

BY: Suvendu Praharaj

We feel cared about when we receive a personalized gift on our birthday or a handwritten thank you note. On the flip side, we might feel someone doesn’t know us that well if we receive a more generic gift that isn’t aligned with our interests or hobbies.

The same can be said for companies that want to deliver a world-class product experience for their consumers. Personalization can be defined as tailoring an experience unique to the individual or a group of customers based on what the products or the companies have learned about their audience.

There are ample examples of how various companies provide a personalized experience to their customers. For instance, the music streaming service Spotify curates personalized playlists based on users’ listening history, most played genres, preferred artists, etc.

In terms of online shopping, brands can suggest related or relevant product recommendations in real time by looking at data points such as previous product purchases or products browsed online in the past.

 

Personalization for construction?

 

But what does personalization mean for the construction industry? Does it even matter for our customers?

Construction projects are inherently different from each other, but still might have some similarities. For example, a rail project is different from an airport project, and an airport project is very different from a high-rise building project.

The projects will have a completely different set of activities and require different machines and equipment, and tradespeople with different skillsets. However, these projects can also face the same risks, such as labor or material shortages.

When machine learning models are used to predict project delays, a one size fits all approach would negatively affect the prediction accuracy, because no two projects are the same.

 

The power of AI-based personalization

 

If that’s the case, how can we ensure that prediction models consider these innate differences while providing predictions? This is where the power of personalization comes into effect.

One organisation might prefer to use a precast flat panel system of construction where most of the work is done offsite and the final structure is erected on site for a high-rise commercial tower. Whereas, another organisation might prefer a flat slab method for a similar high-rise building with greater design flexibility, including curved shapes, ramps, and raised partition walls.

 

Project similarities and differences

 

It’s important to consider the similarities as well as the differences between various construction projects while making predictions.

Our recently launched AI solution, Construction Intelligence Cloud Advisor, uses machine learning to continually analyze project data managed in Oracle Construction and Engineering solutions to identify potential risks and inefficiencies early, empowering organizations to make better decisions. Organizations can start getting project delay predictions using the solution’s off-the-shelf prediction models, irrespective of whether users have a considerable amount of historical data or not.

Fig 1: The Construction Intelligence Cloud Advisor machine learning workbench. Customers use the seed model for delay predictions.

Customers can also retrain Construction Intelligence Cloud Advisor’s prediction models on their own completed past projects to get customized delay predictions. In addition, the models learn as projects progress and more data is created; we term this dynamic learning process “active intelligence.”

Fig 2: The CIC ML workbench where customers use a custom model based on their own data for retraining and delay predictions

 

Creating project groups

 

Customers can create project groups and allocate projects based on different categories, such as industry sectors and segments (including airport, rail, road, residential, etc.) regions (US, Asia, Europe), or based on construction capital. Construction Intelligence Cloud Advisor can use these project groups to make better predictions for similar ongoing projects.

For instance, a general contractor might be executing projects in different regions of the world. The method of construction, project risk, and manpower availability can vary from location to location.

The Oracle predictive solution can create project groups based on locations so that the delay prediction of a US specific project will be based on similar past US projects.

Fig 3:  Custom Project Group within CIC Advisor

 

Project delay predictions

 

Construction Intelligence Cloud Advisor also makes individual activity delay predictions and identifies the key factors responsible for delays as well. For example, the Oracle solution can suggest there is a high likelihood of an electrical commissioning activity getting delayed by 15 days because the resources assigned to these activities have not completed tasks 80% of the time in the past.

However, just as projects are different, so are the activities within the projects. A foundation activity will be entirely different from an electrical commissioning activity. Imagine a less experienced project manager is estimating for a foundation activity.

The project manager would like to get a feel for past durations of similar activities, common risks encountered, and factors that have delayed such activities in the past.

The Oracle predictive tool can help project managers by creating trends and patterns to group similar past activities by using AI and machine learning. The intelligent solution can provide advanced analytics, including: typical delays in the past, teams that successfully executed projects on time, factors that delayed the activities, etc. which help in planning for similar projects.

The next natural evolution

Learn more about Oracle Construction Intelligence Cloud Service.

Download our free report

Predictive artificial intelligence in construction

The art of construction meets the science of data

Download this brief to learn how to put your data to work to mitigate risk and improve decision-making:

  • See how advanced technologies, such as AI, are easily unlocking new predictive insights from project data
  • Learn how to make the most of your growing volume of data to improve on-time, on-budget project delivery
  • Understand the power of active intelligence for construction project management
  • Explore how AI can improve outcomes in schedule, cost/budget, quality, safety, risk, and collaboration management

Privacy Policy.

Additional details regarding Oracle’s collection and use of your personal information,including information about access, retention, rectification, deletion, security, cross-border transfers and other topics, is available in the Oracle Privacy Policy.

Oracle values your privacy and will treat the information we collect from you as a result of your registration in accordance with the Oracle Privacy Policy, which is located here Oracle utilizes a third party registration company manage registration and logistical activities for this program.’