Artificial intelligence in construction is no longer just buzz and hype. It’s a make-or-break business reality. AI leverages computers to do what they do best—and allows people to do the same. While many of the major players are already hyper-aware of AI as a business imperative, and that investing in it will give them a competitive edge, only a handful are ready to introduce the technologies right now.

But first, what is AI? One common definition is that AI is a kind of general intelligence where computational systems emulate the full spectrum of human thinking, behaviors, interactions—and beyond. This kind of AI is a long way off and will likely look a lot different than it appears today.

The second type of AI is a kind of specific intelligence—computational systems that can perform specific, targeted activities (such as recognize an object from an image, or validate an invoice, or drive a vehicle). This AI is built using a toolbox of data science, machine learning and optimization techniques that work together to more effectively solve business problems – and this is the AI that is making a difference in business today. AI is a tool, like other tools, that allows people to accomplish work and achieve objectives with enhanced speed and efficiency.

The next generation of AI tools in the construction industry will free up human resources from the administrative aspects of the work today (particularly activities such as data collection, entry and validation) in order to pursue higher value activities (e.g. asking questions, identifying constraints, prioritizing and decision making).

AI tools are well suited to collect and validate data, identify patterns and anomalous events, and optimize systems to adjust for such patterns or events. As a human, how many dimensions can really be considered simultaneously when performing an analysis? AI tools can quickly tease out patterns and relationships across much higher dimensional systems. They take a wider, holistic view of the organization’s data to do things like identify potentially harmful business situations (e.g., detection of a risk condition) or erroneous process conditions (e.g. an invoice that doesn’t make sense, lacks enough detail and requires review).

AI also brings a new potential for lean programs in managing and reducing waste. Waste is a huge problem in the industry and there are efficiencies to be uncovered throughout the transactional hierarchy. AI allows automation and process controls to play a greater role in the industry and unlocks latent productivity.

The majority of business leaders already understand that AI, machine learning, robotics and IoT are critical for the future. They also know that they can’t boil the ocean. Resources are limited and taking on a new big project is a serious endeavor. A 2018 report by McKinsey revealed that, despite the proven ROI of AI for construction and engineering firms, only a few are currently in a position to implement them. Most lack the personnel, processes and/or tools.

A good initial step is to assess existing data. Investigate how current data sources can be integrated and streamlined. Organizations that take a holistic approach to their data are better positioned to take advantage of advances in AI, robotics and IoT. For companies that aren’t quite ready to implement an AI pilot program, creating a holistic data environment is a starting point.

Another important strategic move is to ensure the organizational model is up to date. Intensely hierarchical organizational structures can make it more difficult to introduce change. Decentralizing a bit can help to ensure that when rolling out new models, team members will embrace the different way of working. Also, keep in mind that where hierarchies can be collapsed, efficiencies can be gained.

Together, this means the organization should:

• examine current business systems for opportunities to break away from any legacy paper-based processes;
• integrate systems so they can talk to each other without wasting human time;
• enable better transparency between teams (teams should be able to work independently with continuous visibility over progress); and
• work on the company’s digital culture adoption—the majority of issues associated to digital transformation come from internal adoption concerns.

If the company is waiting for AI to become more mainstream before moving forward with a major implementation, consider testing the waters sooner than later. Start in an area that’s of strategic importance for the organization. For example, if the organization does a high volume of annual building, or if the business model is to compete on price, it could test partial automation of early-stage feasibility assessments, which tend to be a costly, time-consuming part of the construction process. If HR is an unusually expensive area of the business, look at testing AI’s ability to automate portions of the recruiting and onboarding process. Whatever the use case, establish clear goals and objectives and quarterly milestones to achieve them.

Some of the new AI pilot applications in construction include schedule optimization, team optimization, cost estimation and sensor and camera footage analysis and pattern-finding. While a more comprehensive approach is generally recommended, small-scale pilots are great because they tend to lead to bigger things down the road. A related and important data trend is the ability to quickly scale computing infrastructure. So, as an organization develops and deploys their AI initiatives, it can build on successes and scale to support a wider consumer base (enterprise, vendors, suppliers, etc.) with greater confidence and lower risk.

While hesitation is normal—especially in the ‘early adopters’ stage—it’s important to recognize the following risk/opportunity: companies that leverage AI may have a significant and lasting competitive advantage over those that do not. Think about business strategies and where AI could fit in.