Alliance for artificial intelligence in construction - Kwant

Predicting the Future with Historical Experience. A Research Project from Kwant and GroupPMX

The intent of this post is to illustrate a new way of applying the knowledge gained in past projects to present scenarios, through deep learning and AI. This is an invitation to collaborate, contribute and take advantage of the first kind of effort in the industry.

Why is AI important in construction?

If you look at the progress in construction over the past decade, every process from design to construction has shifted from a tedious paper-based process to a digital format and then to the cloud and mobile.

This has enabled the project team to have more control and visibility over the project and improved productivity via a seamless flow of bytes from field to a fancy dashboard on an executive's desk.

We’ve already been collecting various project data, from schedules to daily reports, to manpower logs, and so on, all of which are stored in siloed software solutions. Since construction data is contextual and its contexts are dynamic in nature, the data that is collected has no useful meaning outside the context of the project.

This makes the insights gained from a project difficult to transfer between projects, or between people and locations, even within the same firm. Another aspect of the problem is that humans are inherently biased by nature. This creates the potential for previous mistakes to be repeated which in turn lowers the productivity and profit margin.

This is the area where deep learning can be used to analyze the data from historical projects to predict disasters & workplace accidents before they even occur. This allows us to dodge mistakes which occur repeatedly in multiple projects. We can see one step ahead, from better control over the project to predicting risks even before they occur.

Why now?

We have seen many other industries powered by data analytics, artificial intelligence, and machine learning. Media, marketing, and social media companies sending targeted advertisements, Google maps optimizing your driving routes, or Alexa finding new music you want to hear.

We are surrounded by virtual assistants helping us save time and be more profitable in our professional as well as personal lives.

‍

It’s time for the construction industry to take the leap. The construction and extraction industry is expected to grow 11% between 2016 and 2026, faster than the 7% average forecast for all occupations.

According to the United Nations, 68% of the world population is projected to live in urban areas by 2050. It has never been more important for our industry to rise above the ever occurring cost overruns, delays, and safety issues.

• We need to improve productivity/reduce waste which translates to better profit margin - For eg: resource optimization and predictive planning using real-time data using mobile apps and the Internet of things
• Predictive bidding, planning, and procurement, where you can prioritize which project has a higher rate of success or which activities have a higher risk of causing project delays.
• Predicting workplace incident rates based on worker movement, behavior, or their location, time, and the type of activity being performed based on historical safety audits.

We are inviting all construction professionals to share their data ( In an anonymous format ) to contribute to this Alliance for Artificial Intelligence in Construction Industry.

Relevant types of data

• Schedules with different updates in Xer ( or any format)
• Daily reports
• Cost Estimates

Recommended Data format

• Xer files :Schedules of different versions( updates) (pdf, mpp, xer or any other file format)
• Daily Reports or timesheets ( in xls, pdf or any other application)

What's in it for you!

Our data science team has already built the first phase of the model for proactive schedule and resource planning. This is already being used on more than 30 projects around the world and providing benefits by increasing productivity, saving time, money, and increasing the safety of these job sites.

• We will offer this version of the platform to you immediately
• We will provide advisory services, and offer key insight by recognizing patterns regarding where you were losing money and which company/crew has higher levels of productivity.

‍